On 15/10/14 19:45, Alexey Indeev wrote:
> Hey,
>
> Pekka, We will definitely do some testing on CentOS and try to reproduce
> these issues. Could you let us know which network driver you are using and
> whether you are running inside a VM or not?
>
Hiya,
virtio_net running on a qemu-kvm 2.1.2 host.
here's the oops (ah, noticed you just found the same bug and filed an
issue):
<6>[quick_tx] INFO: Device registered: /dev/net/quick_tx_lo --> lo
<6>[quick_tx] INFO: Device registered: /dev/net/quick_tx_eth0 --> eth0
<3>kmem_cache_create: duplicate cache skbuff_head_cache
<4>Pid: 23196, comm: insmod Not tainted 2.6.32-431.29.2.el6.x86_64 #1
<4>Call Trace:
<4> [<ffffffff811723a8>] ? kmem_cache_create+0x538/0x5a0
<4> [<ffffffffa034f81e>] ? quick_tx_init+0x4ce/0x650 [quick_tx]
<4> [<ffffffff8103ea6c>] ? kvm_clock_read+0x1c/0x20
<4> [<ffffffffa034f350>] ? quick_tx_init+0x0/0x650 [quick_tx]
<4> [<ffffffff8100204c>] ? do_one_initcall+0x3c/0x1d0
<4> [<ffffffff810bc291>] ? sys_init_module+0xe1/0x250
<4> [<ffffffff8100b072>] ? system_call_fastpath+0x16/0x1b
<0>Kernel panic - not syncing: kmem_cache_create(): failed to create
slab `skbuff_head_cache'
<0>
<4>Pid: 23196, comm: insmod Not tainted 2.6.32-431.29.2.el6.x86_64 #1
<4>Call Trace:
<4> [<ffffffff8152873c>] ? panic+0xa7/0x16f
<4> [<ffffffff81010ee5>] ? show_trace+0x15/0x20
<4> [<ffffffff81172390>] ? kmem_cache_create+0x520/0x5a0
<4> [<ffffffffa034f81e>] ? quick_tx_init+0x4ce/0x650 [quick_tx]
<4> [<ffffffff8103ea6c>] ? kvm_clock_read+0x1c/0x20
<4> [<ffffffffa034f350>] ? quick_tx_init+0x0/0x650 [quick_tx]
<4> [<ffffffff8100204c>] ? do_one_initcall+0x3c/0x1d0
<4> [<ffffffff810bc291>] ? sys_init_module+0xe1/0x250
<4> [<ffffffff8100b072>] ? system_call_fastpath+0x16/0x1b
...
trying to naively fix that by renaming to "qtx_skbuff_head_cache"
makes things happy until actually using it:
<2>kernel BUG at drivers/net/virtio_net.c:576!
<4>invalid opcode: 0000 [#1] SMP
<4>last sysfs file: /sys/module/nfnetlink/initstate
<4>CPU 0
<4>Modules linked in: quick_tx(U) nfnetlink_queue nfnetlink_log
nfnetlink bluetooth rfkill xt_socket nf_conntrack nf_defrag_ipv4
nf_defrag_ipv6 ipv6 xt
_MARK iptable_mangle iptable_filter ip_tables ext3 jbd microcode
virtio_balloon virtio_console snd_hda_intel snd_hda_codec snd_hwdep
snd_seq snd_seq_de
vice snd_pcm snd_timer snd soundcore snd_page_alloc virtio_net i2c_piix4
i2c_core ext4 jbd2 mbcache virtio_blk virtio_pci virtio_ring virtio
pata_acpi
ata_generic ata_piix dm_mirror dm_region_hash dm_log dm_mod [last
unloaded: speedstep_lib]
<4>
<4>Pid: 1466, comm: quick_tx_workqu Not tainted
2.6.32-431.29.2.el6.x86_64 #1 QEMU Standard PC (i440FX + PIIX, 1996)
<4>RIP: 0010:[<ffffffffa01146b1>] [<ffffffffa01146b1>]
xmit_skb+0x251/0x290 [virtio_net]
<4>RSP: 0018:ffff88007b19da20 EFLAGS: 00010246
<4>RAX: 000000000000ff00 RBX: ffff88007b19da30 RCX: ffff88007d682020
<4>RDX: ffff88007d682000 RSI: 0000000000000000 RDI: ffff88007b19dcb0
<4>RBP: ffff88007b19dcd0 R08: 0000000000000000 R09: ffff88007b19da30
<4>R10: 000000000000004a R11: 000000000000002a R12: ffff880068ea2e90
<4>R13: ffff880068ea2ec8 R14: ffff880037c676e0 R15: ffffffffa0392d60
<4>FS: 0000000000000000(0000) GS:ffff880002200000(0000)
knlGS:0000000000000000
<4>CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b
<4>CR2: 0000000001899018 CR3: 0000000037a27000 CR4: 00000000000006f0
<4>DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
<4>DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
<4>Process quick_tx_workqu (pid: 1466, threadinfo ffff88007b19c000, task
ffff88007da9eaa0)
<4>Stack:
<4> 0000000000000082 ffff88007d682020 0000000000000000 0000000000000000
<4><d> 0000000000000000 0000000000000000 0000000000000000 0000000000000000
<4><d> 0000000000000000 0000000000000000 0000000000000000 0000000000000000
<4>Call Trace:
<4> [<ffffffffa0114738>] start_xmit+0x48/0x1d0 [virtio_net]
<4> [<ffffffffa0390a50>] quick_tx_worker+0x4a0/0x1520 [quick_tx]
<4> [<ffffffff8109afa0>] ? autoremove_wake_function+0x0/0x40
<4> [<ffffffffa03905b0>] ? quick_tx_worker+0x0/0x1520 [quick_tx]
<4> [<ffffffff81094a20>] worker_thread+0x170/0x2a0
<4> [<ffffffff8109afa0>] ? autoremove_wake_function+0x0/0x40
<4> [<ffffffff810948b0>] ? worker_thread+0x0/0x2a0
<4> [<ffffffff8109abf6>] kthread+0x96/0xa0
<4> [<ffffffff8100c20a>] child_rip+0xa/0x20
<4> [<ffffffff8109ab60>] ? kthread+0x0/0xa0
<4> [<ffffffff8100c200>] ? child_rip+0x0/0x20
<4>Code: 00 00 00 41 c6 44 24 38 01 66 29 d0 66 41 89 45 06 41 0f b7 44
24 76 66 41 89 45 08 e9 35 fe ff ff 41 c6 45 01 04 e9 8e fe ff ff <0f>
0b eb fe
f6 05 7c 6c ec e1 40 0f 84 f5 fd ff ff 80 3d 76 2e
<1>RIP [<ffffffffa01146b1>] xmit_skb+0x251/0x290 [virtio_net]
<4> RSP <ffff88007b19da20>
which would probably be:
if (skb_is_gso(skb)) {
hdr->hdr.hdr_len = skb_headlen(skb);
hdr->hdr.gso_size = skb_shinfo(skb)->gso_size;
if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV4)
hdr->hdr.gso_type = VIRTIO_NET_HDR_GSO_TCPV4;
else if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6)
hdr->hdr.gso_type = VIRTIO_NET_HDR_GSO_TCPV6;
else if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP)
hdr->hdr.gso_type = VIRTIO_NET_HDR_GSO_UDP;
else
BUG();
reloading virtio_net with gso=0 didn't seem to help (but I should be
going to bed and might have not actually reloaded it properly :P )
------------------------------------------------------------------------------
Comprehensive Server Monitoring with Site24x7.
Monitor 10 servers for $9/Month.
Get alerted through email, SMS, voice calls or mobile push notifications.
Take corrective actions from your mobile device.
http://p.sf.net/sfu/Zoho
_______________________________________________
Tcpreplay-users mailing list
Tcpreplay-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/tcpreplay-users
Support Information: http://tcpreplay.synfin.net/trac/wiki/Support
