kvm vlan virtio problem

Tue Jan 22 10:15:35 UTC 2013

Hi!

The same warning shows up in our setup:

Jan 21 23:40:46 host kernel: WARNING: at net/core/dev.c:1712
skb_gso_segment+0x1df/0x2b0() (Tainted: G        W  ---------------   )
Jan 21 23:40:46 host kernel: Hardware name: System Product Name
Jan 21 23:40:46 host kernel: tun: caps=(0x1b0049, 0x0) len=4452
data_len=4380 ip_summed=0
[...]

KVM host: CentOS 6.3, Linux kernel 2.6.32-279.19.1.el6.x86_64
VM guest: FreeBSD 9.1, virtio-kmod-9.1-0.242658

Disabling TSO on vtnet0 stops the warnings on the KVM host.

Is there any progress on this issue?

	Best regards
		Franz

On 04.11.2012 02:53, Bryan Venteicher wrote:
> Hi,
> 
> ----- Original Message -----
>> From: "Bane Ivosev" <bane.ivosev at pmf.uns.ac.rs>
>> To: "Bryan Venteicher" <bryanv at daemoninthecloset.org>, freebsd-stable at freebsd.org
>> Sent: Saturday, November 3, 2012 7:58:57 PM
>> Subject: Re: kvm vlan virtio problem
>>
>> thanks bryan, i don't have vlans inside guest.
>>
>> as you suggested i disabled nic tso
>>
>> ifconfig vtnet0 -tso
>> and
>> sysctl  net.inet.tcp.tso=0 and
>>
>> so far, so good!
>> everything is ok now.
>>
>> once again, thanks a lot.
>>
> 
> No need to disable TSO globally. So it seems if_vtnet sending
> down TSO frames that don't have the appropriate flags set. I'll
> look into it more in a next couple of days.
> 
> Thanks for the quick reply back.
> 
> Bryan 
> 
>>
>> On 11/03/2012 09:40 PM, Bryan Venteicher wrote:
>>> Hi,
>>>
>>> ----- Original Message -----
>>>> From: "Bane Ivosev"<bane.ivosev at pmf.uns.ac.rs>
>>>> To: freebsd-stable at freebsd.org
>>>> Sent: Saturday, November 3, 2012 2:12:25 PM
>>>> Subject: kvm vlan virtio problem
>>>>
>>>> hi, i have several kvm ubuntu 12.04 and centos 6 hosts with
>>>> standard
>>>> bridged network setup. same problem on each server with freebsd 9
>>>> amd64
>>>> guest and virtio nic: soon after guest start host syslog is
>>>> filling
>>>> with
>>>> this message at very high rate. guest is working without any
>>>> problem.
>>>> with e1000 guest driver eveything is ok.
>>>>
>>>> does enyone have/had same problem?
>>>>
>>>> thanks.
>>>>
>>>
>>> I have a vague recollection of looking at something similar last
>>> year...
>>>
>>> Do you have VLAN configured in the guest as well? What is the
>>> ifconfig
>>> output? Does disabling TSO on the vtnetX device make these messages
>>> go
>>> away?
>>>
>>> Bryan
>>>
>>>>
>>>>    kernel: [2337728.094141] ------------[ cut here ]------------
>>>>    kernel: [2337728.094144] WARNING: at
>>>> /build/buildd/linux-3.2.0/net/core/dev.c:1955
>>>> skb_gso_segment+0x341/0x3b0()
>>>>    kernel: [2337728.094146] Hardware name: System x3550 M3
>>>>    -[7944K3G]-
>>>>    kernel: [2337728.094148] 802.1Q VLAN Support: caps=(0x30195833,
>>>>    0x0)
>>>> len=3196 data_len=0 ip_summed=0
>>>>    kernel: [2337728.094149] Modules linked in: dm_snapshot
>>>> ip6table_filter ip6_tables ebtable_nat ebtables ipt_MASQUERADE
>>>> iptable_nat nf_nat nf_conntrack_ipv4 nf_defrag_ipv4 xt_state
>>>> nf_conntrack ipt_REJECT xt_CHECKSUM iptable_mangle xt_tcpudp
>>>> iptable_filter ip_tables x_tables kvm_intel kvm dm_crypt nfsd nfs
>>>> lockd
>>>> fscache auth_rpcgss nfs_acl sunrpc nls_iso8859_1 nls_cp437 vfat
>>>> fat
>>>> 8021q garp bridge stp serio_raw cdc_ether usbnet vhost_net macvtap
>>>> i7core_edac macvlan shpchp ioatdma edac_core dca tpm_tis lp
>>>> parport
>>>> mac_hid btrfs zlib_deflate libcrc32c usbhid hid megaraid_sas bnx2
>>>>    kernel: [2337728.094177] Pid: 8685, comm: vhost-8683 Tainted: G
>>>>    W    3.2.0-31-generic #50-Ubuntu
>>>>    kernel: [2337728.094179] Call Trace:
>>>>    kernel: [2337728.094180]<IRQ>   [<ffffffff81066d7f>]
>>>> warn_slowpath_common+0x7f/0xc0
>>>>    kernel: [2337728.094185]  [<ffffffff81066e76>]
>>>>    warn_slowpath_fmt+0x46/0x50
>>>>    kernel: [2337728.094188]  [<ffffffff8153f581>]
>>>>    skb_gso_segment+0x341/0x3b0
>>>>    kernel: [2337728.094191]  [<ffffffff81542ee1>]
>>>> dev_hard_start_xmit+0xc1/0x540
>>>>    kernel: [2337728.094196]  [<ffffffffa01c0150>] ?
>>>>    br_flood+0xc0/0xc0
>>>> [bridge]
>>>>    kernel: [2337728.094199]  [<ffffffff8154360a>]
>>>>    dev_queue_xmit+0x2aa/0x420
>>>>    kernel: [2337728.094203]  [<ffffffffa01c01e2>]
>>>> br_dev_queue_push_xmit+0x92/0xd0 [bridge]
>>>>    kernel: [2337728.094208]  [<ffffffffa01c0278>]
>>>> br_forward_finish+0x58/0x60 [bridge]
>>>>    kernel: [2337728.094212]  [<ffffffffa01c042b>]
>>>>    __br_forward+0xab/0xd0
>>>> [bridge]
>>>>    kernel: [2337728.094217]  [<ffffffffa01c04ed>]
>>>>    br_forward+0x5d/0x70
>>>> [bridge]
>>>>    kernel: [2337728.094221]  [<ffffffffa01c11c2>]
>>>> br_handle_frame_finish+0x182/0x2a0 [bridge]
>>>>    kernel: [2337728.094226]  [<ffffffffa01c14a8>]
>>>> br_handle_frame+0x1c8/0x270 [bridge]
>>>>    kernel: [2337728.094231]  [<ffffffffa01c12e0>] ?
>>>> br_handle_frame_finish+0x2a0/0x2a0 [bridge]
>>>>    kernel: [2337728.094234]  [<ffffffff81540892>]
>>>> __netif_receive_skb+0x1e2/0x520
>>>>    kernel: [2337728.094237]  [<ffffffff81540ff1>]
>>>>    process_backlog+0xb1/0x190
>>>>    kernel: [2337728.094240]  [<ffffffff815422e4>]
>>>>    net_rx_action+0x134/0x290
>>>>    kernel: [2337728.094242]  [<ffffffff8165a4fe>] ?
>>>>    _raw_spin_lock+0xe/0x20
>>>>    kernel: [2337728.094245]  [<ffffffff8106e528>]
>>>>    __do_softirq+0xa8/0x210
>>>>    kernel: [2337728.094248]  [<ffffffff81664d6c>]
>>>>    call_softirq+0x1c/0x30
>>>>    kernel: [2337728.094249]<EOI>   [<ffffffff81015305>]
>>>>    do_softirq+0x65/0xa0
>>>>    kernel: [2337728.094254]  [<ffffffff815427c8>]
>>>>    netif_rx_ni+0x28/0x30
>>>>    kernel: [2337728.094258]  [<ffffffff81478ec6>]
>>>>    tun_get_user+0x306/0x4a0
>>>>    kernel: [2337728.094261]  [<ffffffff81479085>]
>>>>    tun_sendmsg+0x25/0x30
>>>>    kernel: [2337728.094265]  [<ffffffffa01b09b6>]
>>>>    handle_tx+0x296/0x530
>>>> [vhost_net]
>>>>    kernel: [2337728.094269]  [<ffffffffa01b0c85>]
>>>> handle_tx_kick+0x15/0x20 [vhost_net]
>>>>    kernel: [2337728.094273]  [<ffffffffa01ade3d>]
>>>>    vhost_worker+0xdd/0x170
>>>> [vhost_net]
>>>>    kernel: [2337728.094276]  [<ffffffffa01add60>] ?
>>>> vhost_set_memory+0x130/0x130 [vhost_net]
>>>>    kernel: [2337728.094279]  [<ffffffff8108a03c>]
>>>>    kthread+0x8c/0xa0
>>>>    kernel: [2337728.094282]  [<ffffffff81664c74>]
>>>> kernel_thread_helper+0x4/0x10
>>>>    kernel: [2337728.094285]  [<ffffffff81089fb0>] ?
>>>> flush_kthread_worker+0xa0/0xa0
>>>>    kernel: [2337728.094287]  [<ffffffff81664c70>] ?
>>>>    gs_change+0x13/0x13
>>>>    kernel: [2337728.094289] ---[ end trace a38cf088269411b3 ]---
>>>>    kernel: [2337728.094731] ------------[ cut here ]------------