[Gems-users] SIGSEGV with WATTCH on Opal init revisited

[Fred A Bower <fredb@xxxxxxxxxxx>]

I've done some further digging into my issue with a SIGSEGV getting thrown 
on Opal init when WATTCH is enabled ( 
https://lists.cs.wisc.edu/archive/gems-users/2008-August/msg00027.shtml ). 
I have tracked the problem down to 
the fact that WATTCH calls out_intf_t::out_info() prior to the cycle 
counter getting initialized.  This causes the segmentation fault when 
out_intf_t::validDebug() gets called and system_t::inst->getGlobalCycle() 
is invoked.  I hotwired this function to always return true and I now get 
into simulation successfully.
I'm not sure if this indicates that I have something wrong in my mfacet.py 
or some other portion of my environment, but it appears that this problem 
would be hard to avoid without some code intervention.
Thanks,
fred
fredb@xxxxxxxxxxx