forrtl: severe (174): SIGSEGV, segmentation fault occurred

General scientific issues regarding ROMS

Moderators: arango, robertson

Post Reply
Message
Author
backkom
Posts: 35
Joined: Wed Nov 07, 2007 5:03 pm
Location: Ocean University of China

forrtl: severe (174): SIGSEGV, segmentation fault occurred

#1 Unread post by backkom »

Hi all,

I want to know if there is anyone else has met the error described as follows:

ROMS Reversion: 667

Operating system : Linux
CPU/hardware : x86_64
Compiler system : ifort
Compiler command : /opt/mpi/intel/mvapich2-1.8.1/bin/mpif90
Compiler flags : -heap-arrays -fp-model precise -ip -O3 -free -free -free

When I test the default ideal upwelling case, the compiling can be done successfully, however, a segmentation fault occurred when I run this oceanM, the detailed information about this error are like this:
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image PC Routine Line Source
oceanM 00000000004AC625 Unknown Unknown Unknown
oceanM 00000000004A1233 Unknown Unknown Unknown
oceanM 0000000000488917 Unknown Unknown Unknown
oceanM 000000000040796C Unknown Unknown Unknown
oceanM 000000000040778B Unknown Unknown Unknown
oceanM 00000000004075DC Unknown Unknown Unknown
libc.so.6 0000003A6801ECDD Unknown Unknown Unknown
oceanM 00000000004074D9 Unknown Unknown Unknown
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image PC Routine Line Source
oceanM 00000000004AC625 Unknown Unknown Unknown
oceanM 00000000004A1233 Unknown Unknown Unknown
oceanM 0000000000488917 Unknown Unknown Unknown
oceanM 000000000040796C Unknown Unknown Unknown
oceanM 000000000040778B Unknown Unknown Unknown
oceanM 00000000004075DC Unknown Unknown Unknown
libc.so.6 0000003A6801ECDD Unknown Unknown Unknown
oceanM 00000000004074D9 Unknown Unknown Unknown


After some tests, I found everything can goes well if the AVERAGE cpp option is not activated. So here I want to know if there is anyone else has similar experience since I am not sure whether it is a bug of ROMS or just some improper enviromental settings for my case?

Best,

User avatar
kate
Posts: 4091
Joined: Wed Jul 02, 2003 5:29 pm
Location: CFOS/UAF, USA

Re: forrtl: severe (174): SIGSEGV, segmentation fault occurr

#2 Unread post by kate »

I have gotten seg faults and there is more than one way to get them. If you want to track it down, the first thing to try is to compile with USE_DEBUG. Then you might still get the seg fault and it would give you line numbers instead of all that useless Unknown garbage. Alternately, it won't seg fault when not optimized, which makes it tough to know if it's a ROMS bug or a compiler bug or something else.

One thing it could be is hitting a memory limit of some kind. Having AVERAGES on will increase your memory needs quite a bit.

Post Reply