LCP插件创建对等物理接口

如下创建物理接口GigabitEthernet7/0/0的对等接口e7。

vpp# show interface GigabitEthernet7/0/0
              Name               Idx    State  MTU (L3/IP4/IP6/MPLS)     Counter          Count
GigabitEthernet7/0/0              10    down         9000/0/0/0
vpp#
vpp# lcp create GigabitEthernet7/0/0 host-if e7
vpp#
vpp# show tap
Interface: tap10 (ifindex 17)
  name "e7"
  host-mtu-size "9000"
  host-mac-addr: 00:60:e0:6f:5a:f0

创建物理接口GigabitEthernet8/0/0的对等接口e8，并且接口状态设置为UP，MTU设置为1500（默认情况为9000）。对等接口的MAC地址与实际的物理接口的MAC地址相同。

vpp# show interface GigabitEthernet8/0/0
              Name               Idx    State  MTU (L3/IP4/IP6/MPLS)     Counter          Count
GigabitEthernet8/0/0              11    down         9000/0/0/0
vpp#
vpp#
vpp# set interface state GigabitEthernet8/0/0 up
vpp# set interface mtu packet 1500 GigabitEthernet8/0/0
vpp#
vpp# lcp create GigabitEthernet8/0/0 host-if e8
vpp#
vpp# show tap
Interface: tap11 (ifindex 18)
  name "e8"
  host-mtu-size "1500"
  host-mac-addr: 00:60:e0:6f:5a:f1

LCP命令查看接口对。

vpp# show lcp
lcp default netns '<unset>'
lcp lcp-auto-subint off
lcp lcp-sync off
itf-pair: [0] GigabitEthernet7/0/0 tap10 e7 18 type tap
itf-pair: [1] GigabitEthernet8/0/0 tap11 e8 19 type tap

退出VPP命令行，在Linux shell中查看创建的两个tap接口，e7的mtu为9000，状态为down。接口e8的mtu为1500，状态为up。并且，接口e7和e8的MAC地址与VPP中的接口GigabitEthernet7/0/0和GigabitEthernet8/0/0的MAC地址相同。

$ ip link show type tun
18: e7: <BROADCAST,MULTICAST> mtu 9000 qdisc mq state DOWN mode DEFAULT group default qlen 1000
    link/ether 00:60:e0:6f:5a:f0 brd ff:ff:ff:ff:ff:ff
19: e8: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UNKNOWN mode DEFAULT group default qlen 1000
    link/ether 00:60:e0:6f:5a:f1 brd ff:ff:ff:ff:ff:ff

管理状态同步

LCP接口管理状态同步开关。

vpp# lcp lcp-sync on
vpp#
vpp# show lcp
lcp lcp-sync on

如下在VPP中将GigabitEthernet7/0/0接口状态设置为UP，linux中的对等接口e7状态也同步为UP。

vpp# set interface state GigabitEthernet7/0/0 up
vpp# quit
#
#
# ip link show type tun
18: e7: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 9000 qdisc mq state DOWN mode DEFAULT group default qlen 1000
    link/ether 00:60:e0:6f:5a:f0 brd ff:ff:ff:ff:ff:ff

接口链路状态不受以上开关的限制。断开物理接口GigabitEthernet8/0/0的网线，VPP和linux中的e8接口的链路状态都是DOWN。

# ip link show type tun
19: e8: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UNKNOWN mode DEFAULT group default qlen 1000
    link/ether 00:60:e0:6f:5a:f1 brd ff:ff:ff:ff:ff:ff
#
#
# vppctl
vpp#
vpp# show hardware-interfaces GigabitEthernet8/0/0
              Name                Idx   Link  Hardware
GigabitEthernet8/0/0              11    down  GigabitEthernet8/0/0

vpp#
vpp# quit
#
# ip link show type tun
19: e8: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN mode DEFAULT group default qlen 1000
    link/ether 00:60:e0:6f:5a:f1 brd ff:ff:ff:ff:ff:ff

设置接口IP地址

VPP中物理接口和linux中对等接口e8设置完全相同的IP地址。

vpp# set interface ip address GigabitEthernet8/0/0 10.0.1.1/24
vpp# quit
vpp#
#
# ip addr add dev e8 10.0.1.1/24

在linux中ping外部主机10.0.1.2。

# ping 10.0.1.2
PING 10.0.1.2 (10.0.1.2) 56(84) bytes of data.
64 bytes from 10.0.1.2: icmp_seq=1 ttl=64 time=0.290 ms

在VPP中trace回复的ICMP报文，如下：

vpp# trace add dpdk-input 100
vpp#
vpp# show trace

Packet 3

00:40:13:066975: dpdk-input
  GigabitEthernet7/0/0 rx queue 0
  buffer 0x82ca6: current data 0, length 98, buffer-pool 0, ref-count 1, trace handle 0x2000002
                  ext-hdr-valid
  PKT MBUF: port 9, nb_segs 1, pkt_len 98
    buf_len 2176, data_len 98, ol_flags 0x180, data_off 128, phys_addr 0x92cb2a00
    packet_type 0x11 l2_len 0 l3_len 0 outer_l2_len 0 outer_l3_len 0
    rss 0x0 fdir.hi 0x0 fdir.lo 0x0
    Packet Offload Flags
      PKT_RX_IP_CKSUM_GOOD (0x0080) IP cksum of RX pkt. is valid
      PKT_RX_IP_CKSUM_NONE (0x0090) no IP cksum of RX pkt.
      PKT_RX_L4_CKSUM_GOOD (0x0100) L4 cksum of RX pkt. is valid
      PKT_RX_L4_CKSUM_NONE (0x0108) no L4 cksum of RX pkt.
    Packet Types
      RTE_PTYPE_L2_ETHER (0x0001) Ethernet packet
      RTE_PTYPE_L3_IPV4 (0x0010) IPv4 packet without extension headers
  IP4: 54:e1:ad:3f:11:f4 -> 00:60:e0:6f:5a:f0
  ICMP: 10.0.1.2 -> 10.0.1.1
    tos 0x00, ttl 128, length 84, checksum 0xd71b dscp CS0 ecn NON_ECN
    fragment id 0x4d8b
  ICMP echo_reply checksum 0xecf9 id 3
00:40:13:066978: ethernet-input
  frame: flags 0x3, hw-if-index 10, sw-if-index 10
  IP4: 54:e1:ad:3f:11:f4 -> 00:60:e0:6f:5a:f0
00:40:13:066980: ip4-input-no-checksum
  ICMP: 10.0.1.2 -> 10.0.1.1
    tos 0x00, ttl 128, length 84, checksum 0xd71b dscp CS0 ecn NON_ECN
    fragment id 0x4d8b
  ICMP echo_reply checksum 0xecf9 id 3
00:40:13:066980: ip4-lookup
  fib 0 dpo-idx 7 flow hash: 0x00000000
  ICMP: 10.0.1.2 -> 10.0.1.1
    tos 0x00, ttl 128, length 84, checksum 0xd71b dscp CS0 ecn NON_ECN
    fragment id 0x4d8b
  ICMP echo_reply checksum 0xecf9 id 3
00:40:13:066984: ip4-receive
    ICMP: 10.0.1.2 -> 10.0.1.1
      tos 0x00, ttl 128, length 84, checksum 0xd71b dscp CS0 ecn NON_ECN
      fragment id 0x4d8b
    ICMP echo_reply checksum 0xecf9 id 3
00:40:13:066985: ip4-icmp-input
  ICMP: 10.0.1.2 -> 10.0.1.1
    tos 0x00, ttl 128, length 84, checksum 0xd71b dscp CS0 ecn NON_ECN
    fragment id 0x4d8b
  ICMP echo_reply checksum 0xecf9 id 3
00:40:13:066985: ip4-icmp-echo-reply
  ICMP4 echo id 3 seq 1 (unknown)
00:40:13:066988: ip4-punt
    ICMP: 10.0.1.2 -> 10.0.1.1
      tos 0x00, ttl 128, length 84, checksum 0xd71b dscp CS0 ecn NON_ECN
      fragment id 0x4d8b
    ICMP echo_reply checksum 0xecf9 id 3
00:40:13:066988: ip4-punt-redirect
  via redirect:0
00:40:13:066989: ip4-dvr-dpo 
     sw_if_index:17
00:40:13:066990: ip4-dvr-reinject
     sw_if_index:17
00:40:13:066992: tap11-output
  tap11
  IP4: 54:e1:ad:3f:11:f4 -> 00:60:e0:6f:5a:f0
  ICMP: 10.0.1.2 -> 10.0.1.1
    tos 0x00, ttl 128, length 84, checksum 0xd71b dscp CS0 ecn NON_ECN
    fragment id 0x4d8b
  ICMP echo_reply checksum 0xecf9 id 3
00:40:13:066993: tap11-tx
    buffer 0x82ca6: current data 0, length 98, buffer-pool 0, ref-count 1, trace handle 0x2000002
                    ext-hdr-valid
                    local dvr l2-hdr-offset 0 l3-hdr-offset 14
  hdr-sz 0 l2-hdr-offset 0 l3-hdr-offset 14 l4-hdr-offset 0 l4-hdr-sz 0
  IP4: 54:e1:ad:3f:11:f4 -> 00:60:e0:6f:5a:f0
  ICMP: 10.0.1.2 -> 10.0.1.1
    tos 0x00, ttl 128, length 84, checksum 0xd71b dscp CS0 ecn NON_ECN
    fragment id 0x4d8b
  ICMP echo_reply checksum 0xecf9 id 3

在接收到ICMP回复报文后，依次经过以下node节点：

ip4-icmp-echo-reply -> ip4-punt -> ip4-punt-redirect -> ip4-dvr-dpo -> ip4-dvr-reinject ->tap11-output -> tap11-tx

通过tap11-tx将报文送至linux中对等体接口。