[ofa-general] having problems with the latest for-2.6.27 tree

Steve Wise swise at opengridcomputing.com
Wed Jul 9 11:17:13 PDT 2008


Hey Roland,

I'm trying out the latest for-2.6.27 tree and I'm seeing hangs doing 
simple rpings over iwarp.  I'm still investigating, but I see this on 
the server side.  Somebody's in a dead lock methinks:

INFO: task iw_cm_wq:3042 blocked for more than 120 seconds.
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
iw_cm_wq      D ffffffff8125a5a0     0  3042      2
 ffff81002e643d50 0000000000000046 0000000000000000 0000000000000000
 ffff81003e51ab20 ffff81003fddd640 ffff81003e51ae70 00000002a039032d
 ffffffffffffffff ffff810033cb1800 0000000000000000 0000000000000000
Call Trace:
 [<ffffffff8124001b>] __mutex_lock_slowpath+0x5c/0x93
 [<ffffffff8123fea0>] mutex_lock+0x2b/0x2f
 [<ffffffffa038e1f3>] :rdma_cm:cma_disable_callback+0x19/0x36
 [<ffffffffa038ec55>] :rdma_cm:cma_iw_handler+0x1f/0x11c
 [<ffffffffa037db84>] :iw_cm:cm_work_handler+0x280/0x3ef
 [<ffffffffa037d904>] :iw_cm:cm_work_handler+0x0/0x3ef
 [<ffffffff8103fc95>] run_workqueue+0x7b/0x103
 [<ffffffff81040537>] worker_thread+0xd5/0xe0
 [<ffffffff81042cc6>] autoremove_wake_function+0x0/0x2e
 [<ffffffff81040462>] worker_thread+0x0/0xe0
 [<ffffffff81042b99>] kthread+0x47/0x76
 [<ffffffff8100cb98>] child_rip+0xa/0x12
 [<ffffffff81042b52>] kthread+0x0/0x76
 [<ffffffff8100cb8e>] child_rip+0x0/0x12

[root at vic9 ~]#





More information about the general mailing list