Segmentation Fault with ifort version 16.0.1

[JannickWeisshaupt]

When I compile exciting with ifort version 16.0.1 the program crashes on runtime for larger systems and or high Z atoms with segmentation fault.

forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image PC Routine Line Source
excitingser 0000000000D16355 Unknown Unknown Unknown
excitingser 0000000000D14117 Unknown Unknown Unknown
excitingser 0000000000CA7794 Unknown Unknown Unknown
excitingser 0000000000CA75A6 Unknown Unknown Unknown
excitingser 0000000000C28719 Unknown Unknown Unknown
excitingser 0000000000C2EE40 Unknown Unknown Unknown
libpthread.so.0 00007F48A826B100 Unknown Unknown Unknown
excitingser 0000000000487D5E Unknown Unknown Unknown
excitingser 00000000004207F9 Unknown Unknown Unknown
excitingser 000000000063A331 Unknown Unknown Unknown
excitingser 00000000009E1A1B Unknown Unknown Unknown
excitingser 0000000000779172 Unknown Unknown Unknown
excitingser 000000000098E7FC Unknown Unknown Unknown
excitingser 0000000000405A5E Unknown Unknown Unknown
libc.so.6 00007F48A7EBCB15 Unknown Unknown Unknown
excitingser 000000000040596D Unknown Unknown Unknown

Smaller systems with low Z atoms work fine, e.g. Diamond, Si...
I tried to add -heap-arrays to the compiler flags but then the compilation fails with a seg fault as well. With gfortran the executable runs fine but uses only one core.

[JannickWeisshaupt]

I found a way to compile it so that it works but it is extremely ugly.
You have to add -heap-arrays to the compiler flags.
Then compile until the compilation crashes with a seg fault.
Remove the -heap-arrays compiler flag.
Resume compilation and let only the f90 file compile that caused the crash. Then abort compilation.
Add the -heap-arrays flag again and resume compilation.
Repeat roughly 10 times.

[AlexBuccheri]

Closing this issue because it's 4 years out of date. In the future, please compile in debug mode to get a proper stack trace and provide relevant inputs.