Pull networking fixes from David Miller: 1) Limit xt_hashlimit hash table size to avoid OOM or hung tasks, from Cong Wang. 2) Fix deadlock in xsk by publishing global consumer pointers when NAPI is finished, from Magnus Karlsson. 3) Set table field properly to RT_TABLE_COMPAT when necessary, from Jethro Beekman. 4) NLA_STRING attributes are not necessary NULL terminated, deal wiht that in IFLA_ALT_IFNAME. From Eric Dumazet. 5) Fix checksum handling in atlantic driver, from Dmitry Bezrukov. 6) Handle mtu==0 devices properly in wireguard, from Jason A. Donenfeld. 7) Fix several lockdep warnings in bonding, from Taehee Yoo. 8) Fix cls_flower port blocking, from Jason Baron. 9) Sanitize internal map names in libbpf, from Toke Høiland-Jørgensen. 10) Fix RDMA race in qede driver, from Michal Kalderon. 11) Fix several false lockdep warnings by adding conditions to list_for_each_entry_rcu(), from Madhuparna Bhowmik. 12) Fix sleep in atomic in mlx5 driver, from Huy Nguyen. 13) Fix potential deadlock in bpf_map_do_batch(), from Yonghong Song. 14) Hey, variables declared in switch statement before any case statements are not initialized. I learn something every day. Get rids of this stuff in several parts of the networking, from Kees Cook. * git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (99 commits) bnxt_en: Issue PCIe FLR in kdump kernel to cleanup pending DMAs. bnxt_en: Improve device shutdown method. net: netlink: cap max groups which will be considered in netlink_bind() net: thunderx: workaround BGX TX Underflow issue ionic: fix fw_status read net: disable BRIDGE_NETFILTER by default net: macb: Properly handle phylink on at91rm9200 s390/qeth: fix off-by-one in RX copybreak check s390/qeth: don't warn for napi with 0 budget s390/qeth: vnicc Fix EOPNOTSUPP precedence openvswitch: Distribute switch variables for initialization net: ip6_gre: Distribute switch variables for initialization net: core: Distribute switch variables for initialization udp: rehash on disconnect net/tls: Fix to avoid gettig invalid tls record bpf: Fix a potential deadlock with bpf_map_do_batch bpf: Do not grab the bucket spinlock by default on htab batch ops ice: Wait for VF to be reset/ready before configuration ice: Don't tell the OS that link is going down ice: Don't reject odd values of usecs set by user ...
Showing the changes to include/linux (4 of 104 files). Show all
Unified · Split
| @@ -688,7 +688,10 @@ struct mlx5_ifc_flow_table_nic_cap_bits { · 687 unmodified lines | ||
| 688 | 688 | u8 nic_rx_multi_path_tirs[0x1]; |
| 689 | 689 | u8 nic_rx_multi_path_tirs_fts[0x1]; |
| 690 | 690 | u8 allow_sniffer_and_nic_rx_shared_tir[0x1]; |
| 691 | − u8 reserved_at_3[0x1d]; | |
| 691 | + u8 reserved_at_3[0x4]; | |
| 692 | + u8 sw_owner_reformat_supported[0x1]; | |
| 693 | + u8 reserved_at_8[0x18]; | |
| 694 | + | |
| 692 | 695 | u8 encap_general_header[0x1]; |
| 693 | 696 | u8 reserved_at_21[0xa]; |
| 694 | 697 | u8 log_max_packet_reformat_context[0x5]; |
| @@ -72,6 +72,8 @@ void netdev_set_default_ethtool_ops(struct net_device *dev, · 71 unmodified lines | ||
| 72 | 72 | #define NET_RX_SUCCESS 0 /* keep 'em coming, baby */ |
| 73 | 73 | #define NET_RX_DROP 1 /* packet dropped */ |
| 74 | 74 | |
| 75 | +#define MAX_NEST_DEV 8 | |
| 76 | + | |
| 75 | 77 | /* |
| 76 | 78 | * Transmit return codes: transmit return codes originate from three different |
| 77 | 79 | * namespaces: |
| @@ -4389,11 +4391,8 @@ void *netdev_lower_get_next(struct net_device *dev, · 4311 unmodified lines | ||
| 4389 | 4391 | ldev; \ |
| 4390 | 4392 | ldev = netdev_lower_get_next(dev, &(iter))) |
| 4391 | 4393 | |
| 4392 | −struct net_device *netdev_all_lower_get_next(struct net_device *dev, | |
| 4394 | +struct net_device *netdev_next_lower_dev_rcu(struct net_device *dev, | |
| 4393 | 4395 | struct list_head **iter); |
| 4394 | −struct net_device *netdev_all_lower_get_next_rcu(struct net_device *dev, | |
| 4395 | − struct list_head **iter); | |
| 4396 | − | |
| 4397 | 4396 | int netdev_walk_all_lower_dev(struct net_device *dev, |
| 4398 | 4397 | int (*fn)(struct net_device *lower_dev, |
| 4399 | 4398 | void *data), |
| @@ -145,6 +145,13 @@ static inline void hlist_nulls_add_tail_rcu(struct hlist_nulls_node *n, · 144 unmodified lines | ||
| 145 | 145 | } |
| 146 | 146 | } |
| 147 | 147 | |
| 148 | +/* after that hlist_nulls_del will work */ | |
| 149 | +static inline void hlist_nulls_add_fake(struct hlist_nulls_node *n) | |
| 150 | +{ | |
| 151 | + n->pprev = &n->next; | |
| 152 | + n->next = (struct hlist_nulls_node *)NULLS_MARKER(NULL); | |
| 153 | +} | |
| 154 | + | |
| 148 | 155 | /** |
| 149 | 156 | * hlist_nulls_for_each_entry_rcu - iterate over rcu list of given type |
| 150 | 157 | * @tpos: the type * to use as a loop cursor. |
| @@ -611,9 +611,15 @@ typedef unsigned char *sk_buff_data_t; · 610 unmodified lines | ||
| 611 | 611 | * @next: Next buffer in list |
| 612 | 612 | * @prev: Previous buffer in list |
| 613 | 613 | * @tstamp: Time we arrived/left |
| 614 | + * @skb_mstamp_ns: (aka @tstamp) earliest departure time; start point | |
| 615 | + * for retransmit timer | |
| 614 | 616 | * @rbnode: RB tree node, alternative to next/prev for netem/tcp |
| 617 | + * @list: queue head | |
| 615 | 618 | * @sk: Socket we are owned by |
| 619 | + * @ip_defrag_offset: (aka @sk) alternate use of @sk, used in | |
| 620 | + * fragmentation management | |
| 616 | 621 | * @dev: Device we arrived on/are leaving by |
| 622 | + * @dev_scratch: (aka @dev) alternate use of @dev when @dev would be %NULL | |
| 617 | 623 | * @cb: Control buffer. Free for use by every layer. Put private vars here |
| 618 | 624 | * @_skb_refdst: destination entry (with norefcount bit) |
| 619 | 625 | * @sp: the security path, used for xfrm |
| @@ -632,6 +638,9 @@ typedef unsigned char *sk_buff_data_t; · 12 unmodified lines | ||
| 632 | 638 | * @pkt_type: Packet class |
| 633 | 639 | * @fclone: skbuff clone status |
| 634 | 640 | * @ipvs_property: skbuff is owned by ipvs |
| 641 | + * @inner_protocol_type: whether the inner protocol is | |
| 642 | + * ENCAP_TYPE_ETHER or ENCAP_TYPE_IPPROTO | |
| 643 | + * @remcsum_offload: remote checksum offload is enabled | |
| 635 | 644 | * @offload_fwd_mark: Packet was L2-forwarded in hardware |
| 636 | 645 | * @offload_l3_fwd_mark: Packet was L3-forwarded in hardware |
| 637 | 646 | * @tc_skip_classify: do not classify packet. set by IFB device |
| @@ -650,6 +659,8 @@ typedef unsigned char *sk_buff_data_t; · 12 unmodified lines | ||
| 650 | 659 | * @tc_index: Traffic control index |
| 651 | 660 | * @hash: the packet hash |
| 652 | 661 | * @queue_mapping: Queue mapping for multiqueue devices |
| 662 | + * @head_frag: skb was allocated from page fragments, | |
| 663 | + * not allocated by kmalloc() or vmalloc(). | |
| 653 | 664 | * @pfmemalloc: skbuff was allocated from PFMEMALLOC reserves |
| 654 | 665 | * @active_extensions: active extensions (skb_ext_id types) |
| 655 | 666 | * @ndisc_nodetype: router type (from link layer) |
| @@ -660,15 +671,28 @@ typedef unsigned char *sk_buff_data_t; · 4 unmodified lines | ||
| 660 | 671 | * @wifi_acked_valid: wifi_acked was set |
| 661 | 672 | * @wifi_acked: whether frame was acked on wifi or not |
| 662 | 673 | * @no_fcs: Request NIC to treat last 4 bytes as Ethernet FCS |
| 674 | + * @encapsulation: indicates the inner headers in the skbuff are valid | |
| 675 | + * @encap_hdr_csum: software checksum is needed | |
| 676 | + * @csum_valid: checksum is already valid | |
| 663 | 677 | * @csum_not_inet: use CRC32c to resolve CHECKSUM_PARTIAL |
| 678 | + * @csum_complete_sw: checksum was completed by software | |
| 679 | + * @csum_level: indicates the number of consecutive checksums found in | |
| 680 | + * the packet minus one that have been verified as | |
| 681 | + * CHECKSUM_UNNECESSARY (max 3) | |
| 664 | 682 | * @dst_pending_confirm: need to confirm neighbour |
| 665 | 683 | * @decrypted: Decrypted SKB |
| 666 | 684 | * @napi_id: id of the NAPI struct this skb came from |
| 685 | + * @sender_cpu: (aka @napi_id) source CPU in XPS | |
| 667 | 686 | * @secmark: security marking |
| 668 | 687 | * @mark: Generic packet mark |
| 688 | + * @reserved_tailroom: (aka @mark) number of bytes of free space available | |
| 689 | + * at the tail of an sk_buff | |
| 690 | + * @vlan_present: VLAN tag is present | |
| 669 | 691 | * @vlan_proto: vlan encapsulation protocol |
| 670 | 692 | * @vlan_tci: vlan tag control information |
| 671 | 693 | * @inner_protocol: Protocol (encapsulation) |
| 694 | + * @inner_ipproto: (aka @inner_protocol) stores ipproto when | |
| 695 | + * skb->inner_protocol_type == ENCAP_TYPE_IPPROTO; | |
| 672 | 696 | * @inner_transport_header: Inner transport layer header (encapsulation) |
| 673 | 697 | * @inner_network_header: Network layer header (encapsulation) |
| 674 | 698 | * @inner_mac_header: Link layer header (encapsulation) |
| @@ -750,7 +774,9 @@ struct sk_buff { · 75 unmodified lines | ||
| 750 | 774 | #endif |
| 751 | 775 | #define CLONED_OFFSET() offsetof(struct sk_buff, __cloned_offset) |
| 752 | 776 | |
| 777 | + /* private: */ | |
| 753 | 778 | __u8 __cloned_offset[0]; |
| 779 | + /* public: */ | |
| 754 | 780 | __u8 cloned:1, |
| 755 | 781 | nohdr:1, |
| 756 | 782 | fclone:2, |
| @@ -775,7 +801,9 @@ struct sk_buff { · 18 unmodified lines | ||
| 775 | 801 | #endif |
| 776 | 802 | #define PKT_TYPE_OFFSET() offsetof(struct sk_buff, __pkt_type_offset) |
| 777 | 803 | |
| 804 | + /* private: */ | |
| 778 | 805 | __u8 __pkt_type_offset[0]; |
| 806 | + /* public: */ | |
| 779 | 807 | __u8 pkt_type:3; |
| 780 | 808 | __u8 ignore_df:1; |
| 781 | 809 | __u8 nf_trace:1; |
| @@ -798,7 +826,9 @@ struct sk_buff { · 16 unmodified lines | ||
| 798 | 826 | #define PKT_VLAN_PRESENT_BIT 0 |
| 799 | 827 | #endif |
| 800 | 828 | #define PKT_VLAN_PRESENT_OFFSET() offsetof(struct sk_buff, __pkt_vlan_present_offset) |
| 829 | + /* private: */ | |
| 801 | 830 | __u8 __pkt_vlan_present_offset[0]; |
| 831 | + /* public: */ | |
| 802 | 832 | __u8 vlan_present:1; |
| 803 | 833 | __u8 csum_complete_sw:1; |
| 804 | 834 | __u8 csum_level:2; |