Hi, I have a Intel(R) Xeon(R) CPU E5-2680 based system with X710 based 10Gbe card. The i40e driver version used is 1.5.16 and firmware version is 5.02. I have observed one crash where the RIP is pointing to secpath_put when the skb->sp is being accessed although there were no IPSec connections established.
[844890.055882] task: ffff88085439b000 ti: ffff8808543a8000 task.ti: ffff8808543a8000 [844890.063509] RIP: 0010:[<ffffffff816d2811>] [<ffffffff816d2811>] skb_release_head_state+0x31/0xf0 [844890.072559] RSP: 0018:ffff88085fc83dd8 EFLAGS: 00010206 [844890.077991] RAX: 0000000000000000 RBX: ffff881011287900 RCX: 0000000000000001 [844890.085270] RDX: 0000000000000500 RSI: 0000000000000000 RDI: 00000000ffffffff [844890.092548] RBP: ffff88085fc83de0 R08: 0000000000000000 R09: 00000001802a0027 [844890.099828] R10: ffffffff81378240 R11: ffffea004044a100 R12: ffff881011287900 [844890.107108] R13: 00000000ffffff55 R14: ffff8807ca20f550 R15: ffff880803e84a00 [844890.114387] FS: 0000000000000000(0000) GS:ffff88085fc80000(0000) knlGS:0000000000000000 [844890.122620] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [844890.128486] CR2: 00000000ffffffff CR3: 0000000001e0d000 CR4: 00000000001407e0 [844890.135765] Stack: [844890.137890] ffff881011287900 ffff88085fc83df8 ffffffff816d3632 ffff8810138e6548 [844890.145517] ffff88085fc83e18 ffffffff816d3791 ffff8810138e6548 ffff8807ca20f550 [844890.153141] ffff88085fc83e28 ffffffff816e4f85 ffff88085fc83eb0 ffffffffa0075790 [844890.160769] Call Trace: [844890.163328] <IRQ> [844890.165352] [844890.166986] [<ffffffff816d3632>] skb_release_all+0x12/0x30 [844890.171289] [<ffffffff816d3791>] consume_skb+0x31/0x80 [844890.176641] [<ffffffff816e4f85>] __dev_kfree_skb_any+0x35/0x40 [844890.182697] [<ffffffffa0075790>] i40e_napi_poll+0x100/0x1000 [i40e] [844890.189180] [<ffffffff816e92e9>] net_rx_action+0x129/0x220 [844890.194880] [<ffffffff8106fdc7>] __do_softirq+0xb7/0x2a0 [844890.200406] [<ffffffff81070105>] irq_exit+0x95/0xa0 [844890.205489] [<ffffffff8180d916>] do_IRQ+0x56/0xe0 [844890.210401] [<ffffffff818035aa>] common_interrupt+0x6a/0x6a [844890.216185] <EOI> [844890.218205] [844890.219840] [<ffffffff8168b146>] ? cpuidle_enter_state+0x46/0xb0 [844890.224663] [<ffffffff8168b142>] ? cpuidle_enter_state+0x42/0xb0 [844890.230883] [<ffffffff8168b26e>] cpuidle_idle_call+0xbe/0x200 [844890.236839] [<ffffffff8100c06e>] arch_cpu_idle+0xe/0x20 [844890.242277] [<ffffffff810b90c5>] cpu_startup_entry+0xc5/0x260 [844890.248236] [<ffffffff81030680>] start_secondary+0x190/0x1e0 [844890.254099] Code: 48 89 e5 53 48 89 fb 48 8b 7f 58 48 85 ff 74 12 40 f6 c7 01 0f 84 a0 00 00 00 48 c7 43 58 00 00 00 00 48 8b 7b 60 48 85 ff 74 05 <f0> ff 0f 74 7a 48 8b 83 80 00 00 00 48 85 c0 74 15 65 8b 14 25 [844890.274478] RIP [<ffffffff816d2811>] skb_release_head_state+0x31/0xf0 [844890.288994] RSP <ffff88085fc83dd8> This is probably an indication of double free of skb leading to a skb corruption. Any pointers would be helpful. Thanks, Chandrasekhar _______________________________________________ E1000-devel mailing list E1000-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/e1000-devel To learn more about Intel® Ethernet, visit http://communities.intel.com/community/wired
