[ofw] FW: g_al_dbg_level

Tzachi Dar tzachid at mellanox.co.il
Tue Oct 20 15:19:23 PDT 2009


Windows 2008 has a mechanism that checks if one is spending too much
time at dispatch level. If it does it creates an assert.

What I believe is happening is that printing makes functions that have
been running for a short time run much longer (you are printing to the
kernel debugger).

If I'm correct than you can probably use g to continue and than ahi (you
might have to repeat this twice). After that the assert will not happen
(until you reboot).

Thanks
Tzachi

-----Original Message-----
From: ofw-bounces at lists.openfabrics.org
[mailto:ofw-bounces at lists.openfabrics.org] On Behalf Of Hefty, Sean
Sent: Tuesday, October 20, 2009 5:46 PM
To: ofw at lists.openfabrics.org
Subject: [ofw] FW: g_al_dbg_level

forwarding to list...

In windbg, I change g_al_dbg_level value from the default 2 to either 4
of 0xf.  Inevitably I run into an Assertion.  Here is a sample of the
assertion I ran into with the dbg level set to 4:

 

<cut>

[AL]__create_send_av(): [AL]destroy_obj(): [AL]al_dev_ioctl():
[AL]al_dev_ioctl(): fffffa800c2951b0(AL_OBJ_TYPE_H_AV)

al_dev_ioctl: buf_size (64) p_buf (fffffa8015757810).

al_dev_ioctl: buf_size (64) p_buf (fffffa80156ea910).

class 129 attr 4352 AV attr: port num 1, sl 0, dlid 65535, path bits
0[AL]ref_al_obj(): [AL]deref_al_obj():
fffffa80158d2d20(AL_OBJ_TYPE_H_QP)

fffffa800c2951b0(AL_OBJ_TYPE_H_AV)

[AL]__create_send_av(): [AL]deref_al_obj(): [AL]ref_al_obj():
fffffa80158d2d20(AL_OBJ_TYPE_H_QP)

fffffa8015892890(AL_OBJ_TYPE_H_QP)

[AL]detach_al_obj(): [AL]al_dev_ioctl():
fffffa800c2951b0(AL_OBJ_TYPE_H_AV) from
fffffa800c29b7b0(AL_OBJ_TYPE_H_PD)

[AL]deref_al_obj(): Assertion failure - code c0000420 (first chance)

nt! ?? ::FNODOBFM::`string'+0x2e75:

fffff800`0186e421 cd2c            int     2Ch

2: kd> kv

Child-SP          RetAddr           : Args to Child
: Call Site

fffffa60`01999bb0 fffff800`018abe7f : fffffa60`01972180
fffffa60`01999c70 000000c3`1c1222cb fffffa60`0197bd40 : nt! ??
::FNODOBFM::`string'+0x2e75

fffffa60`01999bf0 fffff800`018b68a7 : fffffa60`01972180
fffffa60`00000002 00000001`7b78f8b7 00000000`00000000 :
nt!KiSecondaryClockInterrupt+0x11f (TrapFrame @ fffffa60`01999bf0)

fffffa60`01999d80 fffff800`01a83860 : 00000000`00000000
00000000`00000000 00000000`00000000 00000000`00000000 :
nt!KiIdleLoop+0x27

fffffa60`01999db0 00000000`fffffa60 : 00000000`00000000
00000000`00000000 00000000`00000000 00000000`00000000 :
nt!zzz_AsmCodeRange_End+0x4

fffffa60`01975d00 00000000`00000000 : 00000000`00000000
00000000`00000000 00000000`00000000 00680000`00000000 : 0xfffffa60

2: kd> dd ibbus!g_al_dbg_level 

fffffa60`03350328  00000004 80ffffff 80000000 00000001

fffffa60`03350338  0b3b0250 fffffa80 0c29bd60 fffffa80

fffffa60`03350348  0c29bab0 fffffa80 00000000 00000000

fffffa60`03350358  00000000 00000000 007a0078 00000000

fffffa60`03350368  011736b0 fffff880 007a0078 00000000

fffffa60`03350378  0093f6c0 fffff880 0b3b1c90 fffffa80

fffffa60`03350388  0b3b1f00 fffffa80 0c29b010 fffffa80

fffffa60`03350398  00000000 00000000 0c298010 fffffa80

 

Why does turning on ibbus debug messages cause an assertion and how come
the stack trace isn't help?  I don't understand what it means to have
nt! ?? ::FNODOBFM::`string' on the stack.  

 

Thanks in advance!

Usha

 

___________________

Usha Srinivasan

Software Engineer

QLogic Corporation
780 5th Ave, Suite A

King of Prussia, PA 19406
(610) 233-4844

 

_______________________________________________
ofw mailing list
ofw at lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ofw



More information about the ofw mailing list