got a soft lockup when i try to use it

[ltgoter]

I try to use it in cloudlab, , but got a soft lockup. Do you have any idea aout it?
The detail is as follow:

environmont:
infiniswap, i use the lastest master branch version
ubuntu14.04(3.13.0-168-generic)
MLNX_OFED_LINUX-3.3-1.0.4.0-ubuntu14.04-x86_64
docker-ce 17.06.0~ce-0~ubuntu

hardware
i use two m510 nodes, with ConnectX-3.

I do just as the readme.md said, use the script to install and run successfully. and then use a docker to run app. But when i try to run tpcc benchmark in voltdb, the container and node will down. After i change /proc/sys/kernel/softlockup_panic to 1 to make the node reachable, and rerun it with a simple program just allocting mem continuously in container. When it try to use swap, it get the same error again. I use dmesg, and get the follow message:

[ 680.398599] BUG: soft lockup - CPU#6 stuck for 22s! [docker:11971]
[ 680.426590] Modules linked in: infiniswap(OX) ipt_MASQUERADE nf_conntrack_netlink nfnetlink xfrm_user xfrm_algo iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 xt_addrtype iptable_filter ip_tables xt_conntrack x_tables nf_nat nf_conntrack bridge stp llc aufs nfsv3 ipod(OX) ib_iser(OX) iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi gpio_ich x86_pkg_temp_thermal coretemp kvm_intel kvm crct10dif_pclmul crc32_pclmul ghash_clmulni_intel aesni_intel aes_x86_64 lrw gf128mul glue_helper ablk_helper cryptd lpc_ich ipmi_si shpchp nfsd knem(OX) mac_hid wmi acpi_power_meter auth_rpcgss nfs_acl lp parport nfs lockd sunrpc fscache rdma_ucm(OX) ib_ucm(OX) rdma_cm(OX) iw_cm(OX) configfs ib_ipoib(OX) ib_cm(OX) ib_uverbs(OX) ib_umad(OX) mlx5_ib(OX) mlx5_core(OX) mlx4_ib(OX) ib_sa(OX) ib_mad(OX) ib_core(OX) ib_addr(OX) ib_netlink(OX) mlx4_en(OX) vxlan ip_tunnel ptp pps_core mlx4_core(OX) nvme mlx_compat(OX)
[ 680.791448] CPU: 6 PID: 11971 Comm: docker Tainted: G D OX 3.13.0-168-generic #218-Ubuntu
[ 680.833909] Hardware name: HP ProLiant m510 Server Cartridge/ProLiant m510 Server Cartridge, BIOS H05 05/09/2016
[ 680.882638] task: ffff880ffe6fc800 ti: ffff880fdab22000 task.ti: ffff880fdab22000
[ 680.918015] RIP: 0010:[] [] smp_call_function_many+0x28e/0x2f0
[ 680.958596] RSP: 0000:ffff880fdab23c30 EFLAGS: 00000202
[ 680.982393] RAX: 000000000000000d RBX: ffff88107fcd46a8 RCX: ffff88107fdb7ab8
[ 681.014626] RDX: 000000000000000d RSI: 0000000000000100 RDI: 0000000000000000
[ 681.046588] RBP: ffff880fdab23c80 R08: ffff88107fcd4688 R09: 0000000000000004
[ 681.078563] R10: ffff88107fcd4688 R11: 0000000000000000 R12: 0000000000000006
[ 681.110655] R13: 0000010000000006 R14: 000000000000faa0 R15: 000000fc00013b80
[ 681.142801] FS: 00007f4f157fa700(0000) GS:ffff88107fcc0000(0000) knlGS:0000000000000000
[ 681.179177] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 681.205209] CR2: 00007f4f1e5676c0 CR3: 0000001024c26000 CR4: 0000000000360770
[ 681.237166] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 681.269155] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400

[ 681.301400] Stack:
[ 681.310363] ffff88107fcd46a8 0000000000014640 ffff880fdab23c90 ffffffff81060e20
[ 681.343509] 0000010000000001 ffff8810237c33c0 00007f4f1e5686c0 ffff8810237c3100
[ 681.376719] 00007f4f1e5676c0 ffff880fdab5fb38 ffff880fdab23ca8 ffffffff8106103e
[ 681.410197] Call Trace:
[ 681.421135] [] ? do_kernel_range_flush+0x40/0x40
[ 681.449502] [] native_flush_tlb_others+0x2e/0x30
[ 681.477577] [] flush_tlb_mm_range+0x8a/0x120
[ 681.504522] [] ptep_clear_flush+0x53/0x60
[ 681.529867] [] do_wp_page+0x2a5/0x860
[ 681.554043] [] handle_mm_fault+0x6fb/0xfb0
[ 681.579771] [] ? n_tty_read+0x40c/0xc00
[ 681.604430] [] __do_page_fault+0x183/0x570
[ 681.630373] [] ? wake_up_state+0x20/0x20
[ 681.655321] [] do_page_fault+0x1a/0x70
[ 681.680765] [] page_fault+0x28/0x30
[ 681.703836] Code: 3b 05 cf 52 c3 00 89 c2 0f 8d fd fd ff ff 48 98 49 8b 4d 00 48 03 0c c5 40 7c d1 81 f6 41 20 01 74 cb 0f 1f 00 f3 90 f6 41 20 01 <75> f8 eb be 0f b6 4d d0 48 8b 55 c0 44 89 ef 48 8b 75 c8 e8 ca

Hoping for your help,
thanks.

[ltgoter]

It seems some node could not make this module work in cloudlab. I change to use c6220, it could work.