Mellanox Technologies MT27500 双口网卡做bond 失败，报错Segmentation fault

[fishyu-mushroom]

本机共有三张Mellanox网卡 其中一个为双口的网卡，用于内网bond。dpdk0,dpdk1是在一个Mellanox双口网卡上。
网卡信息如下：

运行报错如下：
root@network-10-48-16-2:/home/intsig# dpvs -- -l 0-10 &
current thread affinity is set to FFFFFFFFFFFF
EAL: Detected 48 lcore(s)
EAL: Detected 1 NUMA nodes
EAL: Detected static linkage of DPDK
EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
EAL: Selected IOVA mode 'PA'
EAL: No available hugepages reported in hugepages-1048576kB
EAL: Probing VFIO support...
EAL: VFIO support initialized
EAL: Invalid NUMA socket, default to 0
EAL: Probe PCI driver: mlx5_pci (15b3:1017) device: 0000:01:00.0 (socket 0)
mlx5_pci: No kernel/verbs support for VF LAG bonding found.
common_mlx5: Failed to load driver = mlx5_pci.

EAL: Requested device 0000:01:00.0 cannot be used
EAL: Invalid NUMA socket, default to 0
EAL: Probe PCI driver: mlx5_pci (15b3:1017) device: 0000:01:00.1 (socket 0)
mlx5_pci: No kernel/verbs support for VF LAG bonding found.
common_mlx5: Failed to load driver = mlx5_pci.

EAL: Requested device 0000:01:00.1 cannot be used
EAL: Invalid NUMA socket, default to 0
EAL: Probe PCI driver: mlx5_pci (15b3:1017) device: 0000:02:00.1 (socket 0)
common_mlx5: RTE_MEM is selected.
mlx5_pci: Size 0xFFFF is not power of 2, will be aligned to 0x10000.
EAL: Invalid NUMA socket, default to 0
EAL: Probe PCI driver: mlx5_pci (15b3:1017) device: 0000:43:00.1 (socket 0)
mlx5_pci: Size 0xFFFF is not power of 2, will be aligned to 0x10000.
EAL: No legacy callbacks, legacy socket not created
DPVS: dpvs version: 1.9-6, build on 2024.07.03.18:33:27
DPVS: dpvs-conf-file: /etc/dpvs.conf
DPVS: dpvs-pid-file: /var/run/dpvs.pid
DPVS: dpvs-ipc-file: /var/run/dpvs.ipc
CFG_FILE: Opening configuration file '/etc/dpvs.conf'.
CFG_FILE: log_level = DEBUG
CFG_FILE: kni = on
NETIF: pktpool_size = 524287 (round to 2^n-1)
NETIF: pktpool_cache_size = 256 (round to 2^n)
NETIF: netif device config: dpdk0
NETIF: dpdk0:rx_queue_number = 8
NETIF: dpdk0:nb_rx_desc = 1024 (round to 2^n)
NETIF: dpdk0:rss = all
NETIF: dpdk0:tx_queue_number = 8
NETIF: dpdk0:nb_tx_desc = 1024 (round to 2^n)
NETIF: dpdk0:kni_name = dpdk0.kni
NETIF: netif device config: dpdk1
NETIF: dpdk1:rx_queue_number = 8
NETIF: dpdk1:nb_rx_desc = 1024 (round to 2^n)
NETIF: dpdk1:rss = all
NETIF: dpdk1:tx_queue_number = 8
NETIF: dpdk1:nb_tx_desc = 1024 (round to 2^n)
NETIF: dpdk1:kni_name = dpdk1.kni
NETIF: netif device config: dpdk2
NETIF: dpdk2:rx_queue_number = 8
NETIF: dpdk2:nb_rx_desc = 1024 (round to 2^n)
NETIF: dpdk2:rss = all
NETIF: dpdk2:tx_queue_number = 8
NETIF: dpdk2:nb_tx_desc = 1024 (round to 2^n)
NETIF: dpdk2:kni_name = dpdk2.kni
NETIF: netif device config: dpdk3
NETIF: dpdk3:rx_queue_number = 8
NETIF: dpdk3:nb_rx_desc = 1024 (round to 2^n)
NETIF: dpdk3:rss = all
NETIF: dpdk3:tx_queue_number = 8
NETIF: dpdk3:nb_tx_desc = 1024 (round to 2^n)
NETIF: dpdk3:kni_name = dpdk3.kni
NETIF: netif bonding config: bond0
NETIF: bonding bond0:mode=4
NETIF: bonding bond0:slave0=dpdk0
NETIF: bonding bond0:slave1=dpdk1
NETIF: bonding bond0:primary=dpdk0
NETIF: bonding bond0:kni_name=bond0.kni
NETIF: bonding bond0 options: dedicated_queues=off
NETIF: netif worker config: cpu0
NETIF: cpu0:type = master
NETIF: cpu0:cpu_id = 0
NETIF: netif worker config: cpu1
NETIF: cpu1:type = slave
NETIF: cpu1:cpu_id = 1
NETIF: worker cpu1:bond0 queue config
NETIF: worker cpu1:bond0 rx_queue_id += 0
NETIF: worker cpu1:bond0 tx_queue_id += 0
NETIF: worker cpu1:dpdk2 queue config
NETIF: worker cpu1:dpdk2 rx_queue_id += 0
NETIF: worker cpu1:dpdk2 tx_queue_id += 0
NETIF: worker cpu1:dpdk3 queue config
NETIF: worker cpu1:dpdk3 rx_queue_id += 0
NETIF: worker cpu1:dpdk3 tx_queue_id += 0
DTIMER: sched_interval = 500
NEIGHBOUR: arp_unres_qlen = 128
NEIGHBOUR: arp_reachable_timeout = 60
IPSET: ipset_hash_pool_size = 131072 (round to 2^n-1)
IPV4: ipv4:forwarding = off
IPV4: inet_def_ttl = 64
IP4FRAG: ip4_frag_buckets = 4096
IP4FRAG: ip4_frag_bucket_entries = 16 (round to 2^n)
IP4FRAG: ip4_frag_max_entries = 4096
IP4FRAG: ip4_frag_ttl = 1
IPV6: ipv6:disable = off
IPV6: ipv6:forwarding = off
RT6: route6:method = hlist
RT6: ipv6:route:recycle_time = 10
MSGMGR: msg_ring_size = 4096 (round to 2^n)
MSGMGR: sync_msg_timeout_us = 20000
MSGMGR: priority_level = low
IPVS: conn_pool_size = 2097152 (round to 2^n-1)
IPVS: conn_pool_cache = 256 (round to 2^n)
IPVS: conn_init_timeout = 3
IPVS: uoa_max_trail = 3
IPVS: udp_timeout_oneway = 60
IPVS: udp_timeout_normal = 300
IPVS: udp_timeout_last = 3
IPVS: tcp_timeout_none = 2
IPVS: tcp_timeout_established = 90
IPVS: tcp_timeout_syn_sent = 3
IPVS: tcp_timeout_syn_recv = 30
IPVS: tcp_timeout_fin_wait = 7
IPVS: tcp_timeout_time_wait = 7
IPVS: tcp_timeout_close = 3
IPVS: tcp_timeout_close_wait = 7
IPVS: tcp_timeout_last_ack = 7
IPVS: tcp_timeout_listen = 120
IPVS: tcp_timeout_synack = 30
IPVS: tcp_timeout_last = 2
IPVS: synack_mss = 1452
IPVS: synack_ttl = 63
IPVS: synproxy_synack_options_sack ON
IPVS: close_client_window ON
IPVS: rs_syn_max_retry = 3
IPVS: ack_storm_thresh = 10
IPVS: max_ack_saved = 3
IPVS: synproxy_conn_reuse ON
IPVS: synproxy_conn_reuse: CLOSE
IPVS: synproxy_conn_reuse: TIMEWAIT
SAPOOL: sapool_filter_enable = on
NETIF: Add bonding device "bond0"
mode: 4
primary: dpdk0
numa_node: 0
slaves: dpdk0 dpdk1
bond_ethdev_mode_set(1603) - Using mode 4, it is necessary to do TX burst and RX burst at least every 100ms.
NETIF: created bondig device bond0: mode=4, primary=dpdk0, numa_node=0
NETIF: bonding device port id range: [2, 3)
DTIMER: [01] timer initialized 0x7f4e9b41d600.
DTIMER: [03] timer initialized 0x7f4e9a41b600.
DTIMER: [02] timer initialized 0x7f4e9ac1c600.
DTIMER: [04] timer initialized 0x7f4e99c1a600.
DTIMER: [05] timer initialized 0x7f4e99419600.
DTIMER: [07] timer initialized 0x7f4e93ffd600.
DTIMER: [06] timer initialized 0x7f4e98c18600.
DTIMER: [10] timer initialized 0x7f4e927fa600.
DTIMER: [09] timer initialized 0x7f4e92ffb600.
DTIMER: [08] timer initialized 0x7f4e937fc600.
DTIMER: [00] timer initialized 0x5633b42d5660.
NETIF: LCORE STATUS
enabled: 0 1 2 3 4 5 6 7 8 9 10
disabled: 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

[1]+ Segmentation fault (core dumped)

dpsv config配置信息如下：

global_defs {
log_level DEBUG
!log_file /var/log/dpvs.log
! log_async_mode on
kni on
}

! netif config
netif_defs {
pktpool_size 524287
pktpool_cache 256

<init> device dpdk0 {
    rx {
        queue_number        8
        descriptor_number   1024
        rss                 all
    }
    tx {
        queue_number        8
        descriptor_number   1024
    }
    ! mtu                   1072
    ! promisc_mode
    ! allmulticast
    kni_name                dpdk0.kni
}
<init> device dpdk1 {
    rx {
        queue_number        8
        descriptor_number   1024
        rss                 all
    }
    tx {
        queue_number        8
        descriptor_number   1024
    }
    ! mtu                    1072
    ! promisc_mode
    ! allmulticast
    kni_name                dpdk1.kni
}
    <init> device dpdk2 {
    rx {
        queue_number        8
        descriptor_number   1024
        rss                 all
    }
    tx {
        queue_number        8
        descriptor_number   1024
    }
    ! mtu                   1072
    ! promisc_mode
    ! allmulticast
    kni_name                dpdk2.kni
}
   <init> device dpdk3 {
    rx {
        queue_number        8
        descriptor_number   1024
        rss                 all
    }
    tx {
        queue_number        8
        descriptor_number   1024
    }
    ! mtu                   1072
    ! promisc_mode
    ! allmulticast
    kni_name                dpdk3.kni
}

<init> bonding bond0 {
    mode                    4
    slave                   dpdk0
    slave                   dpdk1
    primary                 dpdk0
    kni_name                bond0.kni
    options                 dedicated_queues=off
}

}

! worker config (lcores)
worker_defs {
worker cpu0 {
type master
cpu_id 0
}

<init> worker cpu1 {
    type    slave
    cpu_id  1
    port    bond0 {
        rx_queue_ids     0
        tx_queue_ids     0
    }
    port    dpdk2 {
        rx_queue_ids     0
        tx_queue_ids     0
    }
    port    dpdk3 {
        rx_queue_ids     0
        tx_queue_ids     0
    }

}

}

! timer config
timer_defs {
# cpu job loops to schedule dpdk timer management
schedule_interval 500
}

! dpvs neighbor config
neigh_defs {
unres_queue_length 128
timeout 60
}

! dpvs ipset config
ipset_defs {
ipset_hash_pool_size 131072
}

! dpvs ipv4 config
ipv4_defs {
forwarding off
default_ttl 64
fragment {
bucket_number 4096
bucket_entries 16
max_entries 4096
ttl 1
}
}

! dpvs ipv6 config
ipv6_defs {
disable off
forwarding off
route6 {
method hlist
recycle_time 10
}
}

! control plane config
ctrl_defs {
lcore_msg {
ring_size 4096
sync_msg_timeout_us 20000
priority_level low
}
}

! ipvs config
ipvs_defs {
conn {
conn_pool_size 2097152
conn_pool_cache 256
conn_init_timeout 3
! expire_quiescent_template
! fast_xmit_close
! redirect off
}

udp {
    ! defence_udp_drop
    uoa_mode        opp
    uoa_max_trail   3
    timeout {
        oneway      60
        normal      300
        last        3
    }
}

tcp {
    ! defence_tcp_drop
    timeout {
        none        2
        established 90
        syn_sent    3
        syn_recv    30
        fin_wait    7
        time_wait   7
        close       3
        close_wait  7
        last_ack    7
        listen      120
        synack      30
        last        2
    }
    synproxy {
        synack_options {
            mss             1452
            ttl             63
            sack
            ! wscale        0
            ! timestamp
        }
        close_client_window
        ! defer_rs_syn
        rs_syn_max_retry    3
        ack_storm_thresh    10
        max_ack_saved       3
        conn_reuse_state {
            close
            time_wait
            ! fin_wait
            ! close_wait
            ! last_ack
       }
    }
}

}

! sa_pool config
sa_pool {
pool_hash_size 16
flow_enable on
}

[fishyu-mushroom]

如果双线网卡只用一个口。

让后dpdk0 dpdk1做bond，就可以正常启动。

想问下是不支持Mellanox Technologies MT27500 双口网卡 做bond吗？

[fishyu-mushroom]

问题原因是： 在系统层对二个端口做了bond，取消系统层的bond就可以了。