[Openib-windows] (no subject)

[Tzachi Dar]

Hi Fab,
 
While running one of the QP test, one machine was hold by the kernel debugger. When it was freed, I got an assert followed by  a system crush with the following call stack.
 
It seems that in the function ibal!__rej_handler the code is ready to a situation in which it will not found p_cep, and it jumps to err2:. 
However the code there says:
 
 err2:
  KeReleaseInStackQueuedSpinLockFromDpcLevel( &hdl );
 err1:
  ib_put_mad( p_mad );
  AL_EXIT( AL_DBG_CM );
}
...
 
I strongly believe that the function meant to return after the AL_EXIT() statment. 
 
What do you think?
 
Thanks
Tzachi
 
 
ChildEBP RetAddr  Args to Child              
f78aa744 8088d754 f772fa40 00000003 bafa84a0 nt!DbgBreakPoint
f78aaa2c bac0a1bf bac09d60 bac09d30 00000373 nt!RtlAssert+0xba
f78aaa70 bac0ba6f 00000000 8941e7d0 00000001 ibal!__process_rej+0xaf [q:\projinf1\trunk\core\al\kernel\al_cm_cep.c @ 883]
f78aaaa8 bac066f3 8941e7d0 89cea528 00001200 ibal!__rej_handler+0x22f [q:\projinf1\trunk\core\al\kernel\al_cm_cep.c @ 1372]
f78aaad0 babd5815 892bd228 ffffffff 89dfaa90 ibal!__cep_mad_recv_cb+0x1d3 [q:\projinf1\trunk\core\al\kernel\al_cm_cep.c @ 1956]
f78aab04 babcb0ac 892bd228 ffffffff 8941e7d0 ibal!__mad_svc_recv_done+0xa55 [q:\projinf1\trunk\core\al\al_mad.c @ 2206]
f78aab64 bac84dee 89ceb7d0 8941e7d0 89ceaac0 ibal!mad_disp_recv_done+0x12ac [q:\projinf1\trunk\core\al\al_mad.c @ 1004]
f78aab90 bac846bd 89cf3440 8941e7d0 00000001 ibal!process_mad_recv+0x31e [q:\projinf1\trunk\core\al\kernel\al_smi.c @ 2284]
f78aac20 bac84062 89cf3440 89c651d8 ffffffff ibal!spl_qp_comp+0x29d [q:\projinf1\trunk\core\al\kernel\al_smi.c @ 2125]
f78aac48 bac0199b 89c651d8 ffffffff 89cf3440 ibal!spl_qp_recv_comp_cb+0x112 [q:\projinf1\trunk\core\al\kernel\al_smi.c @ 1995]
f78aac64 baf84170 89c651d8 f78aac74 00000000 ibal!ci_ca_comp_cb+0x6b [q:\projinf1\trunk\core\al\kernel\al_ci_ca.c @ 323]
f78aac88 bafa5f14 89c8e008 89dade38 85000000 mthca!cq_comp_handler+0xc0 [q:\projinf1\trunk\hw\mthca\kernel\hca_data.c @ 326]
f78aaca0 bafa87c1 89cea528 00000085 f78aacc0 mthca!mthca_cq_completion+0xa4 [q:\projinf1\trunk\hw\mthca\kernel\mthca_cq.c @ 234]
f78aacd4 bafa84d6 89cea528 89ceaa58 89cea528 mthca!mthca_eq_int+0x81 [q:\projinf1\trunk\hw\mthca\kernel\mthca_eq.c @ 328]
f78aacf8 8083eb0f 89ceaac0 89ceaa58 00000000 mthca!mthca_tavor_dpc+0x36 [q:\projinf1\trunk\hw\mthca\kernel\mthca_eq.c @ 455]
f78aad50 8083ac1f 00000000 0000000e 00000000 nt!KiRetireDpcList+0xca
f78aad54 00000000 0000000e 00000000 00000000 nt!KiIdleLoop+0x37

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openfabrics.org/pipermail/ofw/attachments/20060410/d22bf297/attachment.html>