[LU-3191] lctl segfaults while configuring lnet with linux-3.8.x

Thu, 9 Jan 2020 06:33:05 +0000

after modprobing lnet on linux-3.8.x and issuing
#lctl net up
lctl segfaults. and i get trace in dmesg
[ 4196.143398] LNet: HW CPU cores: 4, npartitions: 1
[ 4250.645327] LNet: Added LNI 10.254.254.254@tcp [8/256/0/180]
[ 4250.645497] LNet: Accept secure, port 988
[ 4250.645611] -----------~~[ cut here ]~~-----------
[ 4250.645622] kernel BUG at fs/inode.c:1440!
[ 4250.645630] invalid opcode: 0000 1 SMP
[ 4250.645646] Modules linked in: ksocklnd(O) lnet(O) libcfs(O) nfsd auth_rpcgss nfs_acl lockd sunrpc bnx2 i2c_piix4 k10temp button
[ 4250.645702] CPU 1
[ 4250.645712] Pid: 20857, comm: lctl Tainted: G O 3.8.7-gentoo #2 HP ProLiant DL385 G5
[ 4250.645730] RIP: 0010:[<ffffffff8112aa68>] [<ffffffff8112aa68>] iput+0x158/0x190
[ 4250.645754] RSP: 0018:ffff8801cc207e08 EFLAGS: 00010202
[ 4250.645763] RAX: 0000000000000000 RBX: ffff88020e9e8cb0 RCX: 0000000000000001
[ 4250.645776] RDX: 0000000000000000 RSI: dead000000100100 RDI: ffff88020e9e8cb0
[ 4250.645790] RBP: ffff8801cc207e28 R08: a018000000000000 R09: ffff88022fc952b0
[ 4250.645809] R10: fdd361f657975403 R11: 0000000000000206 R12: ffff88020e9e8cb0
[ 4250.645811] R13: 0000000000000000 R14: 0000000000000000 R15: ffff880226070220
[ 4250.645814] FS: 00007f0a63eb3700(0000) GS:ffff88022fc80000(0000) knlGS:0000000000000000
[ 4250.645817] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[ 4250.645818] CR2: 00007f87dd4fc620 CR3: 00000001f73fe000 CR4: 00000000000007e0
[ 4250.645819] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 4250.645821] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[ 4250.645824] Process lctl (pid: 20857, threadinfo ffff8801cc206000, task ffff880205b012a0)
[ 4250.645825] Stack:
[ 4250.645831] ffff8801d8976900 ffff88020e9e8cb0 0000000000000000 0000000000000000
[ 4250.645834] ffff8801cc207e58 ffffffff81126b28 ffff88020e9e8c80 ffff8801d8976900
[ 4250.645837] ffff88020e9e8cb0 ffff8801d897695c ffff8801cc207e88 ffffffff811271a9
[ 4250.645838] Call Trace:
[ 4250.645844] [<ffffffff81126b28>] d_kill+0xf8/0x130
[ 4250.645847] [<ffffffff811271a9>] dput+0xc9/0x180
[ 4250.645853] [<ffffffff81112eb2>] __fput+0x122/0x230
[ 4250.645858] [<ffffffff81113079>] ____fput+0x9/0x10
[ 4250.645863] [<ffffffff81055187>] task_work_run+0x97/0xd0
[ 4250.645867] [<ffffffff81002816>] do_notify_resume+0x66/0x70
[ 4250.645871] [<ffffffff8171665a>] int_signal+0x12/0x17
[ 4250.645902] Code: 00 00 eb a1 0f 1f 80 00 00 00 00 48 81 8b a0 00 00 00 00 01 00 00 48 89 df e8 15 fe ff ff 66 83 83 88 00 00 00 01 e9 ef fe ff ff <0f> 0b be 75 05 00 00 48 c7 c7 0e bb 86 81 e8 c5 05 f1 ff e9 05
[ 4250.645906] RIP [<ffffffff8112aa68>] iput+0x158/0x190
[ 4250.645907] RSP <ffff8801cc207e08>
[ 4250.646626] --~~[ end trace a46b6307d03ec7df ]~~--

Second attempt of lctl net up sucsess

Whamcloud Community JIRA

[LU-3191] lctl segfaults while configuring lnet with linux-3.8.x