[openib-general] IBAT kernel oops with latest 3432 svn drop....
Arlin Davis
ardavis at ichips.intel.com
Wed Sep 14 13:04:31 PDT 2005
Arlin Davis wrote:
> Hal,
>
> Can you take a look at this?
>
> My SM was down (ports in INIT state) and I started a uDAPL (dtest -s)
> test that called ib_at_ips_by_gid() and then called ib_at_cancel()
> when it did not complete the request. My application exited with the
> expected open error and then I saw the kernel panic.
>
Question: What can I expect as a result of ib_at_cancel()? Will I always
get an event with -EINTR for the cancelled request id?
> Thanks,
>
> -arlin
>
> Sep 14 12:05:12 iclust-20 start_udev: Starting udev: failed
> Sep 14 12:05:12 iclust-20 kernel: ib_at: ib_dev_ats_op: dev
> (ffffffff8807c620) ib0 already has pending op 2
> Sep 14 12:05:38 iclust-20 kernel: ib_at: req_end: pend
> ffff81003d473e40 already completed? status 0
> Sep 14 12:05:38 iclust-20 kernel: ib_at: req_free: bad async req type 0
> Sep 14 12:05:51 iclust-20 kernel: general protection fault: 0000 [1] SMP
> Sep 14 12:05:51 iclust-20 kernel: CPU 1
> Sep 14 12:05:51 iclust-20 kernel: Modules linked in: ib_uat ib_at
> ib_ucm ib_cm ib_umad ib_uverbs ib_ipoib ib_sa det i b_mthca ib_mad
> ib_core
> Sep 14 12:05:51 iclust-20 kernel: Pid: 2249, comm: ib_at_wq/1 Tainted:
> P 2.6.13
> Sep 14 12:05:51 iclust-20 kernel: RIP: 0010:[<ffffffff8015b524>]
> <ffffffff8015b524>{kfree+168}
> Sep 14 12:05:51 iclust-20 kernel: RSP: 0018:ffff810036b1be18 EFLAGS:
> 00010003
> Sep 14 12:05:51 iclust-20 kernel: RAX: 015492e2f29413d0 RBX:
> ffff81003e108e40 RCX: ffff81000000c000
> Sep 14 12:05:51 iclust-20 kernel: RDX: 002aa23c5e32827a RSI:
> 0000000000000002 RDI: 6172542073736572
> Sep 14 12:05:51 iclust-20 kernel: RBP: 6172542073736572 R08:
> ffff810036b1a000 R09: 0000000000000000
> Sep 14 12:05:51 iclust-20 kernel: R10: 0000000000000001 R11:
> ffffffff805250f8 R12: 0000000000000002
> Sep 14 12:05:51 iclust-20 kernel: R13: 0000000000000292 R14:
> ffff81003e108e40 R15: ffffffff88078a19
> Sep 14 12:05:51 iclust-20 kernel: FS: 0000000000000000(0000)
> GS:ffffffff80597880(0000) knlGS:0000000000000000
> Sep 14 12:05:51 iclust-20 kernel: CS: 0010 DS: 0018 ES: 0018 CR0:
> 000000008005003b
> Sep 14 12:05:51 iclust-20 kernel: CR2: 00002aaaaaaac000 CR3:
> 000000003708f000 CR4: 00000000000006e0
> Sep 14 12:05:51 iclust-20 kernel: Process ib_at_wq/1 (pid: 2249,
> threadinfo ffff810036b1a000, task ffff810036af8730)
> Sep 14 12:05:51 iclust-20 kernel: Stack: 0000000000000206
> ffff81003e108e40 ffff81003c3beb80 ffffffff8807e2a6
> Sep 14 12:05:51 iclust-20 kernel: ffff81003e108e40
> ffff81003e108e68 ffff81003a5e8880 ffffffff88078a34
> Sep 14 12:05:51 iclust-20 kernel: ffff81003e108e60
> ffffffff801427f6
> Sep 14 12:05:51 iclust-20 kernel: Call
> Trace:<ffffffff8807e2a6>{:ib_uat:ib_uat_callback+46}
> <ffffffff88078a34>{:ib_at :req_comp_work+27}
> Sep 14 12:05:51 iclust-20 kernel:
> <ffffffff801427f6>{worker_thread+503}
> <ffffffff8012e864>{default_wake_functi on+0}
> Sep 14 12:05:51 iclust-20 kernel:
> <ffffffff8012e8b2>{__wake_up_common+64}
> <ffffffff8012e864>{default_wake_func tion+0}
> Sep 14 12:05:51 iclust-20 kernel:
> <ffffffff80146620>{keventd_create_kthread+0}
> <ffffffff801425ff>{worker_threa d+0}
> Sep 14 12:05:51 iclust-20 kernel:
> <ffffffff80146620>{keventd_create_kthread+0}
> <ffffffff801465f3>{kthread+204}
> Sep 14 12:05:51 iclust-20 kernel:
> <ffffffff8010e232>{child_rip+8}
> <ffffffff80146620>{keventd_create_kthread+0}
> Sep 14 12:05:51 iclust-20 kernel: <ffffffff80146527>{kthread+0}
> <ffffffff8010e22a>{child_rip+0}
> Sep 14 12:05:51 iclust-20 kernel:
> Sep 14 12:05:51 iclust-20 kernel:
> Sep 14 12:05:51 iclust-20 kernel: Code: 48 8b 78 28 65 8b 04 25 34 00
> 00 00 48 98 48 8b 1c c7 8b 03
> Sep 14 12:05:51 iclust-20 kernel: RIP <ffffffff8015b524>{kfree+168}
> RSP <ffff810036b1be18>
>
> _______________________________________________
> openib-general mailing list
> openib-general at openib.org
> http://openib.org/mailman/listinfo/openib-general
>
> To unsubscribe, please visit
> http://openib.org/mailman/listinfo/openib-general
>
More information about the general
mailing list