Finding Run-time Error

[Intel_C_Intel]

Hello all,

I have an app that compiles & runs fine using CVF, and I am trying to compile & run it w/ IFC on RH8.

It throws up an *Address Error* that I am having trouble fixing. I have recompiled with -C but I can't figure out how to get the debugger to stop _at the error_ in the code. I'm using DDD to debug using the command
ddd -ladebug -debugger idb a.out
as suggested elsewhere in this thread. The app runs, but dies with a run-time error. However, DDD doesn't show me or stop where the error occurs.

Any thoughts?

Matt.

[Brandon_H_Intel]

I've dealt with "*Address Errors*" more than a few times in the past, and every one behaves the same in the debugger:


>run
*Address Error*
>where


Is the problem that the process doesn't seem to stop in idb, so that you can't ask for a stack trace? Or are you able to get a stack trace, but you think it's past the point where the error occurred?

Brandon
Intel Developer Support

For on-line assistance: http://support.intel.com/support/performancetools
For product support information: http://www.intel.com/software/products/support
* Intel and Pentium are registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries
* Other names and brands may be claimed as the property of others

[mbkennel]

Might you try increasing the diagnostic level
with the '-dN' option, e.g. -d2?

Then maybe the runtime error message, absent any debugger, willl still give you the actual line number.

On the other hand I've generally found that when I have a nasty run-time bug I switch back to my NAG compiler, which is very reliable with excellent diagnostics. Its code generaiton is not as good as Intel's though.