[ofw] crash in ipoib build 2223

[Anatoly Greenblatt]

Hi,

 

I got this crash today, the analysis below. I ran opensm then
ibv_rdma_bw -c. Don't know when exactly this happened - I was working
with other node and when I got back the system was already rebooted.

 

Regards,

Anatoly.

 

 

 

1: kd> !analyze  -v

************************************************************************
*******

*
*

*                        Bugcheck Analysis
*

*
*

************************************************************************
*******

 

DRIVER_IRQL_NOT_LESS_OR_EQUAL (d1)

An attempt was made to access a pageable (or completely invalid) address
at an

interrupt request level (IRQL) that is too high.  This is usually

caused by drivers using improper addresses.

If kernel debugger is available get stack backtrace.

Arguments:

Arg1: 00000004, memory referenced

Arg2: 00000002, IRQL

Arg3: 00000001, value 0 = read operation, 1 = write operation

Arg4: 959cb260, address which referenced memory

 

Debugging Details:

------------------

 

 

WRITE_ADDRESS:  00000004 

 

CURRENT_IRQL:  2

 

FAULTING_IP: 

ipoib!cl_qmap_remove_item+2e [c:\work\winof.trunk\core\complib\cl_map.c
@ 1005]

959cb260 895004          mov     dword ptr [eax+4],edx

 

DEFAULT_BUCKET_ID:  VISTA_DRIVER_FAULT

 

BUGCHECK_STR:  0xD1

 

PROCESS_NAME:  System

 

TRAP_FRAME:  805f6748 -- (.trap 0xffffffff805f6748)

ErrCode = 00000002

eax=00000000 ebx=85d24230 ecx=85d24250 edx=00000000 esi=9c625618
edi=9c625618

eip=959cb260 esp=805f67bc ebp=805f67c8 iopl=0         nv up ei pl zr na
pe nc

cs=0008  ss=0010  ds=0023  es=0023  fs=0030  gs=0000
efl=00010246

ipoib!cl_qmap_remove_item+0x2e:

959cb260 895004          mov     dword ptr [eax+4],edx
ds:0023:00000004=????????

Resetting default scope

 

LAST_CONTROL_TRANSFER:  from 959cb260 to 81c75d24

 

STACK_TEXT:  

805f6748 959cb260 badb0d00 00000000 9597b2f3 nt!KiTrap0E+0x2ac

805f67c8 959c2814 85d24230 9c625618 9c625696
ipoib!cl_qmap_remove_item+0x2e
[c:\work\winof.trunk\core\complib\cl_map.c @ 1005]

805f67d8 959c4f64 9c625598 86a20ee8 85e23d08
ipoib!__endpt_mgr_remove+0x56
[c:\work\winof.trunk\ulp\ipoib\kernel\ipoib_port.c @ 4701]

805f6838 959c6515 85d24008 805f6894 85e23d7a ipoib!__recv_arp+0x16e
[c:\work\winof.trunk\ulp\ipoib\kernel\ipoib_port.c @ 2436]

805f6874 959c723a 85d24008 805f6c24 805f6c14
ipoib!__recv_mgr_filter+0x28f
[c:\work\winof.trunk\ulp\ipoib\kernel\ipoib_port.c @ 2115]

805f6c40 83193e17 85c99248 85d24008 805f6c64 ipoib!__recv_cb+0xd2
[c:\work\winof.trunk\ulp\ipoib\kernel\ipoib_port.c @ 1691]

805f6c50 9597af54 85c99248 88000000 85cb52d0 ibbus!ci_ca_comp_cb+0x29
[c:\work\winverbs\core\al\kernel\al_ci_ca.c @ 337]

805f6c64 9597dab4 85d436e0 00000088 805dcb00
mthca!mthca_cq_completion+0xd4
[c:\work\winof.trunk\hw\mthca\kernel\mthca_cq.c @ 245]

805f6cc0 9597d976 85d436e0 85d43c70 85d436e0 mthca!mthca_eq_int+0x94
[c:\work\winof.trunk\hw\mthca\kernel\mthca_eq.c @ 329]

805f6ce8 81cd1450 85d43cf0 85d43c70 00000000 mthca!mthca_tavor_dpc+0x36
[c:\work\winof.trunk\hw\mthca\kernel\mthca_eq.c @ 466]

805f6d50 81ccfedd 00000000 0000000e 00000000 nt!KiRetireDpcList+0x147

805f6d54 00000000 0000000e 00000000 00000000 nt!KiIdleLoop+0x49

 

 

STACK_COMMAND:  kb

 

FOLLOWUP_IP: 

ipoib!cl_qmap_remove_item+2e [c:\work\winof.trunk\core\complib\cl_map.c
@ 1005]

959cb260 895004          mov     dword ptr [eax+4],edx

 

FAULTING_SOURCE_CODE:  

  1001:              CL_ASSERT( p_del_item != &p_map->nil );

  1002: }

  1003: 

  1004: /* Remove the item from the list. */

> 1005:             __cl_primitive_remove( &p_item->pool_item.list_item
);

  1006: /* Decrement the item count. */

  1007: p_map->count--;

  1008: 

  1009: /* Get the pointer to the new root's child, if any. */

  1010: if( p_del_item->p_left != &p_map->nil )

 

 

SYMBOL_STACK_INDEX:  1

 

SYMBOL_NAME:  ipoib!cl_qmap_remove_item+2e

 

FOLLOWUP_NAME:  MachineOwner

 

MODULE_NAME: ipoib

 

IMAGE_NAME:  ipoib.sys

 

DEBUG_FLR_IMAGE_TIMESTAMP:  4a2286e2

 

FAILURE_BUCKET_ID:  0xD1_ipoib!cl_qmap_remove_item+2e

 

BUCKET_ID:  0xD1_ipoib!cl_qmap_remove_item+2e

 

Followup: MachineOwner

---------

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openfabrics.org/pipermail/ofw/attachments/20090601/4b4e7f4c/attachment.html>