Bug 42693 - WARNING: at net/core/dev.c:1886 skb_gso_segment+0x2ab/0x320()
Summary: WARNING: at net/core/dev.c:1886 skb_gso_segment+0x2ab/0x320()
Status: NEEDINFO
Alias: None
Product: Networking
Classification: Unclassified
Component: IPV4 (show other bugs)
Hardware: All Linux
: P1 high
Assignee: Stephen Hemminger
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2012-01-30 12:28 UTC by Igor Novgorodov
Modified: 2020-03-27 11:46 UTC (History)
2 users (show)

See Also:
Kernel Version: 3.0.18
Subsystem:
Regression: Yes
Bisected commit-id:


Attachments

Description Igor Novgorodov 2012-01-30 12:28:09 UTC
These warnings pops up a few times a second when a relatively high load is imposed on 10Gbit ethernet, and causing performance degradation.
Looking into code, it seems to be related to segmentation offload, GSO, and checksums.
The igb and ixgbe modules are out-of-kernel versions from sourceforge.
IGB 3.3.6 and IXGBE 3.7.17, latest at the moment.
When i've fallen back to the 2.6.39.4 with the same modules, the problems disappear.

[  826.097369] ------------[ cut here ]------------
[  826.097372] WARNING: at net/core/dev.c:1886 skb_gso_segment+0x2ab/0x320()
[  826.097374] ixgbe: caps=(0x72114bb3, 0x0) len=3150 data_len=0 ip_summed=0
[  826.097375] Modules linked in: ixgbe igb
[  826.097377] Pid: 0, comm: swapper Tainted: G        W   3.0.18 #2
[  826.097379] Call Trace:
[  826.097379]  <IRQ>  [<ffffffff8104162c>] ? warn_slowpath_common+0x6c/0xa0
[  826.097383]  [<ffffffff81041715>] ? warn_slowpath_fmt+0x45/0x50
[  826.097386]  [<ffffffff8132353b>] ? skb_gso_segment+0x2ab/0x320
[  826.097389]  [<ffffffff81324f1e>] ? dev_hard_start_xmit+0x21e/0x5f0
[  826.097391]  [<ffffffff8133af53>] ? sch_direct_xmit+0xf3/0x1c0
[  826.097393]  [<ffffffff81325445>] ? dev_queue_xmit+0x155/0x5a0
[  826.097395]  [<ffffffff81274b0a>] ? bond_start_xmit+0x3da/0x470
[  826.097397]  [<ffffffff81366e38>] ? tcp_v4_rcv+0x4c8/0x740
[  826.097400]  [<ffffffff81324f62>] ? dev_hard_start_xmit+0x262/0x5f0
[  826.097402]  [<ffffffff81345db5>] ? ip_local_deliver+0x85/0x110
[  826.097404]  [<ffffffff81325696>] ? dev_queue_xmit+0x3a6/0x5a0
[  826.097406]  [<ffffffff8138a53d>] ? br_dev_queue_push_xmit+0x4d/0x70
[  826.097408]  [<ffffffff8138b11b>] ? br_handle_frame_finish+0x10b/0x1c0
[  826.097410]  [<ffffffff8138b371>] ? br_handle_frame+0x1a1/0x210
[  826.097412]  [<ffffffff81274510>] ? bond_set_carrier+0xa0/0xa0
[  826.097414]  [<ffffffff8138b1d0>] ? br_handle_frame_finish+0x1c0/0x1c0
[  826.097416]  [<ffffffff81322d8d>] ? __netif_receive_skb+0x12d/0x370
[  826.097419]  [<ffffffff81323e48>] ? netif_receive_skb+0x58/0x80
[  826.097421]  [<ffffffff81324249>] ? dev_gro_receive+0x1c9/0x2c0
[  826.097423]  [<ffffffff81324943>] ? napi_gro_receive+0xa3/0xc0
[  826.097427]  [<ffffffffa0064108>] ? ixgbe_poll+0x3f8/0x13f0 [ixgbe]
[  826.097429]  [<ffffffff81070800>] ? hotplug_cfd+0x10/0x40
[  826.097431]  [<ffffffff81324541>] ? net_rx_action+0x81/0x120
[  826.097433]  [<ffffffff81046f78>] ? __do_softirq+0x98/0x120
[  826.097435]  [<ffffffff8107ed2a>] ? handle_irq_event_percpu+0x7a/0x140
[  826.097437]  [<ffffffff813a228c>] ? call_softirq+0x1c/0x30
[  826.097439]  [<ffffffff81003f2d>] ? do_softirq+0x4d/0x80
[  826.097441]  [<ffffffff81047266>] ? irq_exit+0x96/0xc0
[  826.097443]  [<ffffffff81003dec>] ? do_IRQ+0x5c/0xd0
[  826.097445]  [<ffffffff813a0d93>] ? common_interrupt+0x13/0x13
[  826.097446]  <EOI>  [<ffffffff81009738>] ? mwait_idle+0x58/0x80
[  826.097450]  [<ffffffff810009ec>] ? cpu_idle+0x4c/0x90
[  826.097452]  [<ffffffff8154485b>] ? start_kernel+0x2b0/0x2bb
[  826.097453] ---[ end trace f7478101639067dd ]---
Comment 1 Stephen Hemminger 2020-03-27 03:05:30 UTC
Please reproduce with current kernel.
Comment 2 Igor Novgorodov 2020-03-27 11:46:08 UTC
Lets close this old one maybe.
Most probably not valid anymore.

Note You need to log in before you can comment on or make changes to this bug.