#2 igb_poll Bug?

closed
igb (34)
in-kernel_driver
5
2015-01-31
2012-03-23
liu dalin
No

In my linux 2.6.38.8, a igb_poll caused the oops:

[147330.974222] Pid: 0, comm: kworker/0:1 Tainted: P W 2.6.38.8 #195 To be filled by O.E.M. To be filled by O.E.M./P8B-X series
[147331.117915] EIP: 0060:[<c01c2353>] EFLAGS: 00010202 CPU: 5
[147331.184371] EIP is at put_page+0x3/0x40
[147331.231127] EAX: b6a7662e EBX: 00000001 ECX: ee19b0e0 EDX: b6a7662e
[147331.306914] ESI: ecaab840 EDI: f3de0860 EBP: f3cffe64 ESP: f3cffe54
[147331.382700] DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068
[147331.448119] Process kworker/0:1 (pid: 0, ti=f3cfe000 task=f3cd4b00 task.ti=f3ce6000)
[147331.541531] Stack:
[147331.566515] f3cffe64 c062b424 ecaab840 ecaab840 f3cffe70 c062afe2 ecaab840 f3cffe80
[147331.660755] c062b0af ecaab840 00000000 f3cffe8c c0634137 f8868108 f3cffea8 f827ac53
[147331.754996] 00000246 f3cffea8 f8868108 0000000b f3570500 f3cfff90 f827ca91 f3cffec4
[147331.849238] Call Trace:
[147331.879409] [<c062b424>] ? skb_release_data+0x84/0xa0
[147331.941717] [<c062afe2>] kfree_skb+0x12/0x90
[147331.996768]
[<c062b0af>] consume_skb+0x4f/0x70
[147332.051819]
[<c0634137>] dev_kfree_skb_any+0x37/0x40
[147332.113094]
[<f827ac53>] igb_unmap_and_free_tx_resource+0x53/0x80 [igb]
[147332.194064]
[<f827ca91>] igb_poll+0xd1/0xd00 [igb]
[147332.253260]
[<c0167014>] ?
hrtimer_start_range_ns+0x154/0x400
[147332.325938] [<c013fdc9>] ? rebalance_domains+0x89/0x150
[147332.390319] [<c063763a>] net_rx_action+0xaa/0x1a0
[147332.448481] [<c014bfa1>] do_softirq+0xb1/0x190
[147332.505606]
[<c014bef0>] ?
do_softirq+0x0/0x190
[147332.563767] <IRQ>
[147332.589894] [<c014be6d>] ? irq_exit+0x5d/0x80
[147332.643907] [<c0104bc5>] ? do_IRQ+0x45/0xb0
[147332.695850] [<c01092d8>] ? sched_clock+0x8/0x10
[147332.751938] [<c01036f0>] ? common_interrupt+0x30/0x38
[147332.814245] [<c016007b>] ? find_ge_pid+0x1b/0x40
[147332.871373] [<c0396c31>] ? acpi_idle_enter_bm+0x256/0x28a
[147332.937827] [<c0605469>] ? cpuidle_idle_call+0xd9/0x1c0
[147333.002208] [<c010214a>] ? cpu_idle+0x8a/0xc0
[147333.056226] [<c077fe49>] ? start_secondary+0x1a1/0x1e8
[147333.119568] Code: 04 f0 ff 0e 0f 94 c0 84 c0 74 d4 89 f8 e8 96 fe ff ff eb cb 0f ae e8 89 f6 8b 03 eb de 8d 74 26 00 8d bc 27 00 00 00 00 55 89 c2 <66> f7 00 00 c0 89 e5 75 1d 8b 40 04 f0 ff 4a 04 0f 94 c0 84 c0

////////////////////////////////////////////////////
igb driver is in-kernel version:

ethtool -i eth2

driver: igb
version: 2.1.0-k2
////////////////////////////////////////////////////
Before this problem occurs, the kernel on skb_gso_segment the WARNNING:

[147320.465292] ------------[ cut here ]------------
[147320.465297] WARNING: at net/core/dev.c:1861 skb_gso_segment+0x225/0x300()
[147320.465298] Hardware name: To be filled by O.E.M.
[147320.465299] : caps=(0x2000, 0x0) len=78 data_len=0 ip_summed=1
[147320.465300] Modules linked in: nf_connmark ip_set_hash_net ip_set_hash_ip xt_set ip_set xt_hashrate xt_pcc xt_nth xt_random(P) xt_nflog xt_replace xt_index igb
[147320.465306] Pid: 0, comm: kworker/0:1 Tainted: P 2.6.38.8 #195
[147320.465307] Call Trace:
[147320.465310] [<c0635415>] ? skb_gso_segment+0x225/0x300
[147320.465312] [<c0145eac>] ? warn_slowpath_common+0x6c/0xa0
[147320.465314] [<c0635415>] ? skb_gso_segment+0x225/0x300
[147320.465315] [<c0145f5e>] ? warn_slowpath_fmt+0x2e/0x30
[147320.465317] [<c0635415>] ? skb_gso_segment+0x225/0x300
[147320.465319] [<c063570f>] ? dev_hard_start_xmit+0x21f/0x5d0
[147320.465321] [<c016cd59>] ? ktime_get+0x69/0xf0
[147320.465325] [<c064b57d>] ? sch_direct_xmit+0x13d/0x1a0
[147320.465327] [<c064b690>] ? qdisc_run+0xb0/0x100
[147320.465329]
[<c0632aaf>] ? net_tx_action+0x11f/0x180
[147320.465332]
[<c014bfa1>] ?
do_softirq+0xb1/0x190
[147320.465334] [<c014bef0>] ? do_softirq+0x0/0x190
[147320.465335] <IRQ>
[<c014be6d>] ? irq_exit+0x5d/0x80
[147320.465339]
[<c011c9a6>] ? smp_apic_timer_interrupt+0x56/0x90
[147320.465341]
[<c0168d4d>] ? sched_clock_cpu+0x12d/0x190
[147320.465344]
[<c0786411>] ? apic_timer_interrupt+0x31/0x38
[147320.465346]
[<c016007b>] ? find_ge_pid+0x1b/0x40
[147320.465349]
[<c0396d53>] ? acpi_idle_enter_simple+0xee/0x122
[147320.465352]
[<c0605469>] ? cpuidle_idle_call+0xd9/0x1c0
[147320.465353]
[<c010214a>] ? cpu_idle+0x8a/0xc0
[147320.465356]
[<c077fe49>] ? start_secondary+0x1a1/0x1e8
[147320.465358] ---[ end trace 45142ab47417677f ]---
[147320.590386] ------------[ cut here ]------------
[147320.590390] WARNING: at net/core/dev.c:1861 skb_gso_segment+0x225/0x300()
[147320.590391] Hardware name: To be filled by O.E.M.
[147320.590393] : caps=(0x2000, 0x0) len=41 data_len=0 ip_summed=1
[147320.590393] Modules linked in: nf_connmark ip_set_hash_net ip_set_hash_ip xt_set ip_set xt_hashrate xt_pcc xt_nth xt_random(P) xt_nflog xt_replace xt_index igb
[147320.590399] Pid: 28, comm: ksoftirqd/7 Tainted: P W 2.6.38.8 #195
[147320.590399] Call Trace:
[147320.590402]
[<c0635415>] ? skb_gso_segment+0x225/0x300
[147320.590404]
[<c0145eac>] ? warn_slowpath_common+0x6c/0xa0
[147320.590405]
[<c0635415>] ? skb_gso_segment+0x225/0x300
[147320.590407]
[<c0145f5e>] ? warn_slowpath_fmt+0x2e/0x30
[147320.590408]
[<c0635415>] ? skb_gso_segment+0x225/0x300
[147320.590411]
[<c063570f>] ? dev_hard_start_xmit+0x21f/0x5d0
[147320.590412]
[<c016cd59>] ? ktime_get+0x69/0xf0
[147320.590415]
[<c064b57d>] ? sch_direct_xmit+0x13d/0x1a0
[147320.590417]
[<c064b690>] ?
qdisc_run+0xb0/0x100
[147320.590419] [<c0632aaf>] ? net_tx_action+0x11f/0x180
[147320.590422] [<c014bfa1>] ? do_softirq+0xb1/0x190
[147320.590424]
[<c014bef0>] ?
do_softirq+0x0/0x190
[147320.590424] <IRQ> [<c014bb6d>] ? run_ksoftirqd+0x7d/0x120
[147320.590428] [<c014baf0>] ? run_ksoftirqd+0x0/0x120
[147320.590430] [<c0162654>] ? kthread+0x74/0x80
[147320.590431] [<c01625e0>] ? kthread+0x0/0x80
[147320.590433] [<c01036fe>] ? kernel_thread_helper+0x6/0x18
[147320.590434] ---[ end trace 45142ab474176780 ]---
[147320.979664] ------------[ cut here ]------------
[147320.979670] WARNING: at net/core/dev.c:1861 skb_gso_segment+0x225/0x300()
[147320.979671] Hardware name: To be filled by O.E.M.
[147320.979673] : caps=(0x2000, 0x0) len=84 data_len=0 ip_summed=1
[147320.979674] Modules linked in: nf_connmark ip_set_hash_net ip_set_hash_ip xt_set ip_set xt_hashrate xt_pcc xt_nth xt_random(P) xt_nflog xt_replace xt_index igb
[147320.979682] Pid: 25, comm: ksoftirqd/6 Tainted: P W 2.6.38.8 #195
[147320.979683] Call Trace:
[147320.979686] [<c0635415>] ? skb_gso_segment+0x225/0x300
[147320.979689] [<c0145eac>] ? warn_slowpath_common+0x6c/0xa0
[147320.979691] [<c0635415>] ? skb_gso_segment+0x225/0x300
[147320.979693] [<c0145f5e>] ? warn_slowpath_fmt+0x2e/0x30
[147320.979695] [<c0635415>] ? skb_gso_segment+0x225/0x300
……

Related

Bugs: #364
Bugs: #366

Discussion

  • Tushar Dave

    Tushar Dave - 2012-04-03

    Does this problem occur with standalone sourceforge driver?
    How fast does the problem occur? Any specific application/test causing this issue?

     
  • liu dalin

    liu dalin - 2012-04-04

    igb driver is in-kernel version:

    ethtool -i eth2

    driver: igb
    version: 2.1.0-k2

    I think this might be the kernel of GSO Bug, an error is passed to the igb driver of the SKB, resulting in igb_poll in free SKB fails throwing an Oops.

     
  • Tushar Dave

    Tushar Dave - 2012-04-04

    Do you see this issue from other devices , other than igb?
    Try disabling gso with ethtool -K ethx gso off
    Also please provide output of ethtool -k ethx where ethx is interface having issue.

     
  • Tushar Dave

    Tushar Dave - 2012-05-07

    Changing issue status to 'pending'. This will auto close issue in 60 days.
    If you still have issue please update this bug item and we will look into it.

    Thanks.

     
  • liu dalin

    liu dalin - 2012-05-08

    Anshion>./ethtool -k eth2
    Offload parameters for eth2:
    rx-checksumming: off
    tx-checksumming: on
    scatter-gather: off
    tcp-segmentation-offload: off
    udp-fragmentation-offload: off
    generic-segmentation-offload: off
    generic-receive-offload: off
    large-receive-offload: off
    rx-vlan-offload: off
    tx-vlan-offload: off
    ntuple-filters: off
    receive-hashing: on

     
  • Tushar Dave

    Tushar Dave - 2012-08-23

    Sorry for the delay. Do you still have this issue?

     
  • Todd Fujinaka

    Todd Fujinaka - 2013-07-02
    • status: open --> pending
    • assigned_to: Tushar Dave --> Todd Fujinaka
     
  • Todd Fujinaka

    Todd Fujinaka - 2013-07-08
    • status: pending --> closed
     
  • Todd Fujinaka

    Todd Fujinaka - 2013-07-08

    Closing due to inactivity.

     

Log in to post a comment.

Get latest updates about Open Source Projects, Conferences and News.

Sign up for the SourceForge newsletter:





No, thanks