Segmentation fault running SPEC CPU2006

[Zhen_C_]

I use Intel Parallel Studio XE version 16.0.1 to build and run SPEC CPU2006 on Linux, but get segmentation fault when running 410.bwaves.

410.bwaves: copy 0 non-zero return code (exit code=46, signal=0)

****************************************
Contents of bwaves.err
****************************************
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image              PC                Routine            Line        Source
bwaves_base.cpu20  000000000048CD45  Unknown               Unknown  Unknown
bwaves_base.cpu20  000000000048AB07  Unknown               Unknown  Unknown
bwaves_base.cpu20  000000000045E094  Unknown               Unknown  Unknown
bwaves_base.cpu20  000000000045DEA6  Unknown               Unknown  Unknown
bwaves_base.cpu20  0000000000443C56  Unknown               Unknown  Unknown
bwaves_base.cpu20  000000000040E910  Unknown               Unknown  Unknown
libpthread.so.0    00007FC026018BB0  Unknown               Unknown  Unknown
bwaves_base.cpu20  000000000040ACA0  Unknown               Unknown  Unknown
bwaves_base.cpu20  00000000004064D4  Unknown               Unknown  Unknown
bwaves_base.cpu20  000000000040275E  Unknown               Unknown  Unknown
libc.so.6          00007FC025C62DE5  Unknown               Unknown  Unknown
bwaves_base.cpu20  0000000000402669  Unknown               Unknown  Unknown

****************************************

Compiler options:

FC=ifort -m64

OPTIMIZE=-O3

Linux version:

$uname -a

Linux haswell 3.11.0-12-generic #19-Ubuntu SMP Wed Oct 9 16:20:46 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux

[TimP]

https://software.intel.com/en-us/articles/determining-root-cause-of-sigsegv-or-sigbus-errors

suggests the usual possibilities.  In the case of cpu2006, don't the required docs for official submissions show settings including stack limits? -traceback should be used when planning to quote stack dumps. 

I thought an effort was being made to make -heap-arrays safe to use even with auto-parallel.

[KitturGanesh]

Hi Zhen,
Since you're using the Fortran Compiler I'll transfer this to the Fortran Forum, fyi.
_Kittur

[Steven_L_Intel1]

I think I'd go with Tim's suggestion of setting a larger stack size. (limit stacksize unlimited or ulimit stack, depending on shell).

I assume Tim meant -traceback

[TimP]

Amazing what corrections the Android spell checker will make.

[jimdempseyatthecove]

>>Amazing what corrections the Android spell checker will make

This reminds me of my son that lives in Japan. In Japan they have a service for the bar crowd. When you think you have too much to drink, you call the service and two guys show up on a scooter. One of them drives you home. I forget what the name of this service is but my son used it. While waiting for the scooter to show up he texted his girlfriend that he was waiting for the service to pick him up. The autocorrect on his phone changed the service name to "drunken sailors".

Jim Dempsey