Oops - here we go again.
Now it hit another IOC - No special hardware on that one, just the CAN
fieldbus support that has been running on all our IOCs from the beginning.
The only thing that just the two faulty IOCs have in common and the only
thing quite new on them is that the sequencer is loaded and there are seq
state machines running.
-> i
NAME ENTRY TID PRI STATUS PC SP ERRNO DELAY
---------- ------------ -------- --- ---------- -------- -------- ------- -----
...
scanOnce 6fba0e 520a70 65 SUSPEND 6ec252 520a28 0 0
-> checkStack
NAME ENTRY TID SIZE CUR HIGH MARGIN
------------ ------------ -------- ----- ----- ----- ------
...
scanOnce 0x00006fba0e 520a70 9664 72 1076 8588
-> tt "scanOnce"
79c3c _vxTaskEntry +10 : 6fba0e (0, 0, 0, 0, 0, 0, 0, 0, 0, 0)
6fba92 _scanOnce +13e: _dbScanLock (c40064dd)
value = 0 = 0x0
-> l
6fba76 487a ff69 PEA (0xff69,PC)
6fba7a 4878 01c9 PEA 0x000001c9
6fba7e 487a fd7f PEA (0xfd7f,PC)
6fba82 42a7 CLR .L -(A7)
6fba84 61ff 0001 6e64 BSR _errPrintf
6fba8a 504f ADDQ .W #0x8,A7
6fba8c 504f ADDQ .W #0x8,A7
6fba8e 2f2e fffc MOVE .L (0xfffc,A6),-(A7)
6fba92 61ff ffff 07b2 BSR _dbScanLock
6fba98 2eae fffc MOVE .L (0xfffc,A6),(A7)
value = 7322268 = 0x6fba9c = _scanOnce + 0x148
-> l
6fba9c 61ff ffff 1890 BSR _dbProcess
6fbaa2 2eae fffc MOVE .L (0xfffc,A6),(A7)
6fbaa6 61ff ffff 0860 BSR _dbScanUnlock
6fbaac 584f ADDQ .W #0x4,A7
6fbaae 6094 BRA 0x006fba44
6fbab0 4e71 NOP
_scanOnceSetQueueSize:
6fbab2 4e56 0000 LINK .W A6,#0
6fbab6 23ee 0008 0072 9188 MOVE .L (0x8,A6),_onceQueueSize
6fbabe 4280 CLR .L D0
6fbac0 4e5e UNLK A6
value = 7322306 = 0x6fbac2 = _scanOnceSetQueueSize + 0x10
Am I interpreting this right thinking that the reason for the scanOnce
task's crash is its dbScanLock call (first level call caused by a CA wakeup
to the scanOnce task as opposed to subsequent calls from faulty support of
other records in the line of forward linked processing)? I.e. the task
reads junk from the ring buffer? The ring buffer that officially only CA
writes into?
Even more clueless, (I didn't think this to be increasable...)
Ralph
- Replies:
- Re: IOC hangs (still) Ralph Lange
- Re: IOC hangs (still) Ralph Lange
- References:
- Re: IOC hangs (scanOnce crashes) Ralph Lange
- Navigate by Date:
- Prev:
Re: IOC hangs (scanOnce crashes) Ralph Lange
- Next:
EPICS Software Distribution back for business Bakul Banerjee
- Index:
1994
1995
1996
<1997>
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
- Navigate by Thread:
- Prev:
Re: IOC hangs (scanOnce crashes) Ralph Lange
- Next:
Re: IOC hangs (still) Ralph Lange
- Index:
1994
1995
1996
<1997>
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
|