On 17/06/2016 1:23 PM, David Crayford wrote:
On 17/06/2016 1:12 PM, Jack J. Woehr wrote:
Charles Mills wrote:
But there are no explicit calls to either abort() or
terminate() anywhere in the source code.
abort() is called on an unhandled exception.
http://www.cplusplus.com/reference/exception/set_terminate/
That's the most likely cause, most probably thrown by the standard
library. It's not a bad idea to setup your own set_terminate handler
with a ctrace() to
generate a stack trace to see where that's happening.
That's not good advice. abort() will dump and give you a better result.
My theory was that the handler would catch the thrower and create a
stack trace. It doesn't!
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
