WARNING: at net/core/dev.c:1712 skb_gso_segment+0x1df/0x2b0() (Tainted: G W --------------- T)

[hon-y_leong]

I'm running VirtualBox-4.2.14 on RHEL 6.3. I had a VM which uses bridged networking. When I start the VM, the kernel generates a lot of logs, extraction given below:

kernel: ------------[ cut here ]------------

kernel: WARNING: at net/core/dev.c:1712 skb_gso_segment+0x1df/0x2b0() (Tainted: G W --------------- T)

kernel: Hardware name: ProLiant DL360p Gen8

kernel: bnx2x: caps=(0x19c9b3, 0x0) len=1636 data_len=0 ip_summed=1

kernel: Modules linked in: vboxpci(U) vboxnetadp(U) vboxnetflt(U) vboxdrv(U) nfs lockd nfs_acl auth_rpcgss ipt_LOG xt_limit nf_conntrack_ipv4 nf_defrag_ipv4 xt_state nf_conntrack iptable_filter ip_tables autofs4 sunrpc ipmi_devintf ipmi_si ipmi_msghandler pcc_cpufreq cachefiles fscache(T) 8021q garp stp llc vhost_net macvtap macvlan tun kvm_intel kvm uinput power_meter hpilo hpwdt sg microcode serio_raw iTCO_wdt iTCO_vendor_support ioatdma dca shpchp ext4 mbcache jbd2 sd_mod crc_t10dif sr_mod cdrom hpsa(U) bnx2x(U) libcrc32c pata_acpi ata_generic ata_piix dm_mirror dm_region_hash dm_log dm_mod be2iscsi(U) uio cxgb4i cxgb4 cxgb3i libcxgbi cxgb3 mdio libiscsi_tcp libiscsi qla4xxx(U) scsi_transport_iscsi [last unloaded: vboxdrv]

kernel: Pid: 0, comm: swapper Tainted: G W --------------- T 2.6.32-279.el6.x86_64 #1

kernel: Call Trace:

kernel: <IRQ> [<ffffffff8106b747>] ? warn_slowpath_common+0x87/0xc0

kernel: [<ffffffff8106b836>] ? warn_slowpath_fmt+0x46/0x50

kernel: [<ffffffffa02682d5>] ? bnx2x_get_drvinfo+0xb5/0xd0 [bnx2x]

kernel: [<ffffffff8143ad3f>] ? skb_gso_segment+0x1df/0x2b0

kernel: [<ffffffffa04e4662>] ? vboxNetFltLinuxPacketHandler+0x352/0x620 [vboxnetflt]

kernel: [<ffffffff81012bd9>] ? read_tsc+0x9/0x20

kernel: [<ffffffff8109c9a0>] ? getnstimeofday+0x60/0xf0

kernel: [<ffffffff8143a787>] ? netif_receive_skb+0x467/0x6f0

kernel: [<ffffffff8143ca38>] ? netif_receive_skb+0x58/0x60

kernel: [<ffffffff8143cb40>] ? napi_skb_finish+0x50/0x70

kernel: [<ffffffff8143f079>] ? napi_gro_receive+0x39/0x50

kernel: [<ffffffffa0247af7>] ? bnx2x_rx_int+0xbd7/0x16b0 [bnx2x]

kernel: [<ffffffff81019476>] ? tboot_sleep+0x156/0x1e0

kernel: [<ffffffff810137f3>] ? native_sched_clock+0x13/0x80

kernel: [<ffffffff810edfc2>] ? ring_buffer_lock_reserve+0xa2/0x160

kernel: [<ffffffffa024867b>] ? bnx2x_poll+0xab/0x300 [bnx2x]

kernel: [<ffffffff8143f193>] ? net_rx_action+0x103/0x2f0

kernel: [<ffffffff81073ec1>] ? do_softirq+0xc1/0x1e0

kernel: [<ffffffff810db896>] ? handle_IRQ_event+0xf6/0x170

kernel: [<ffffffff8100c24c>] ? call_softirq+0x1c/0x30

kernel: [<ffffffff8100de85>] ? do_softirq+0x65/0xa0

kernel: [<ffffffff81073ca5>] ? irq_exit+0x85/0x90

kernel: [<ffffffff81505af5>] ? do_IRQ+0x75/0xf0

kernel: [<ffffffff8100ba53>] ? ret_from_intr+0x0/0x11

kernel: <EOI> [<ffffffff812cd8ce>] ? intel_idle+0xde/0x170

kernel: [<ffffffff812cd8b1>] ? intel_idle+0xc1/0x170

kernel: [<ffffffff81407627>] ? cpuidle_idle_call+0xa7/0x140

kernel: [<ffffffff81009e06>] ? cpu_idle+0xb6/0x110

kernel: [<ffffffff814f6cdf>] ? start_secondary+0x22a/0x26d

kernel: ---[ end trace 78509b75b9f07003 ]---

The messages kept on repeating themselves. I thought this was fixed in https://www.virtualbox.org/ticket/6872 for VirtualBox-3.x.x. Apparently the problems occurs in VirtualBox-4.x.x.

Anyone encounter similar issues? Appreciate if there is a solution/workaround to resolve this problem.

[Aleksey Ilyushin]

Can you try to either disable GRO with "sudo ethtool <iface> gro off" on the host, or enable it in the guests?

[hon-y_leong]

Replying to aleksey:

Can you try to either disable GRO with "sudo ethtool <iface> gro off" on the host, or enable it in the guests?

Nope, it didn't work. By the way, I run the same VirtualBox version on a server with Intel network adapter, and it turns out fine. The log flooding only happens on server with Broadcom network adapter.

[aeichner]

Please reopen if still relevant with a recent VirtualBox release.