Experimental Physics and
Industrial Control System

On 9/22/21 12:40 AM, Till Straumann wrote:
> Second that. I don't know if you are using run-time linking in which case you would have
> to translate the addresses before decoding them.
> 
> You often also get valuable info from the register contents by correlating them
> with variables (disassemble the code around the fault location; sometimes
> you can find NULL or otherwise fishy pointers...)

Heinz followed up off-list.  It looks like this was the symptom of an issue with epicsEventTrigger()
for RTEMS 5 using POSIX functions which aren't safe for ISRs.

https://github.com/epics-base/epics-base/issues/202


> HTH
> - Till
> 
> On 9/21/21 5:30 PM, Michael Davidsaver via Tech-talk wrote:
>> On 9/21/21 3:03 AM, Heinz Junkes via Tech-talk wrote:
>>> ...
>>> I don't know the best way to resolve the problem (debug, how?).
>>> Setting CASDEBUG outputs too much and I can not cope with this.
>> One place to start is decoding the stack trace using the addr2line utility.
>>
>> ...
>>> Stack Trace:
>>>    IP: 0x00230314, LR: 0x0022a680
>>> --^ 0x0022a67c--^ 0x0014ceb0--^ 0x0014d0ec--^ 0x0014d144--^ 0x00149df8
>>> --^ 0x00142e50--^ 0x001217c0--^ 0x0011a704--^ 0x00231b64--^ 0x0022b0ec
>>> --^ 0x002272c4fatal source: RTEMS_FATAL_SOURCE_EXCEPTION
>> Run:
>>
>>> powerpc-rtems5-addr2line -e ./myexe 0x00230314 0x0022a680 0x0022a67c \
>>>   0x0014ceb0 0x0014d0ec 0x0014d144 0x00149df8 0x00142e50 0x001217c0 \
>>>   0x0011a704 0x00231b64 0x0022b0ec
> 

References:

crash in CAS-event (high load?) Heinz Junkes via Tech-talk

Re: crash in CAS-event (high load?) Michael Davidsaver via Tech-talk

Re: crash in CAS-event (high load?) Till Straumann via Tech-talk

Navigate by Date:

Prev: Re: Quartz crystal microbalance readout SQC-310c device support Yendell, Gary (DLSLtd, RAL, LSCI) via Tech-talk

Next: Ava in caQtDM Mezger Anton Christian (PSI) via Tech-talk

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: crash in CAS-event (high load?) Till Straumann via Tech-talk

Next: modbusTCP liuxiaomin via Tech-talk

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