[CPMD-list] Compilation problem: Intel Fortran 9.1 and em64t

Vladimir Stegailov stegailov at ihed.ras.ru
Fri Jun 8 12:27:35 CEST 2007


Dear Nisanth,
thanks a lot. It works.
(However compilation in the debug mode woth "-g" still fails at the same 
point,
but I think it is of minor importance).
Best regards,
Vladimir


----- Original Message ----- 
From: "Nisanth Nair" <nisanth.nair at theochem.ruhr-uni-bochum.de>
To: <cpmd-list at cpmd.org>
Sent: Friday, June 08, 2007 1:07 PM
Subject: Re: [CPMD-list] Compilation problem: Intel Fortran 9.1 and em64t


> Dear Vladimir,
> Please add -DPOINTER8 to the CPPFLAGS to cure this problem.
> with best regards,
> Nisanth
>
>
> On Friday 08 June 2007 10:14, Vladimir Stegailov wrote:
>> Dear colleagues,
>>
>> could you give some hints concerning the compilation problem.
>>
>> I am trying to compile the code on the Xeon based system with IFC 9.1 and
>> MKL 9.0.
>>
>> The Makefile options are
>>
>> FFLAGS = -c -r8 -w95 -O2 -pc64 -tpp7 -unroll -cm -tune pn4 -arch pn4
>> LFLAGS = -L/opt/intel/mkl/9.0/lib/em64t -lmkl_lapack64 -lmkl_em64t
>> CFLAGS = -c -O2 -Wall
>> CPP = /lib/cpp -P -C -traditional
>> CPPFLAGS = -D__Linux -D__PGI -DFFT_DEFAULT \
>>                -DPARALLEL -DMYRINET -DLINUX_IFC -DINTEL_MKL
>> NOOPT_FLAG =
>> CC = mpicc
>> FC = mpif77
>> LD = mpif77
>>
>> The executable is compiled successfully. However it can not start even
>> with example scripts.
>>
>> Console report is like follows
>>
>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>> Image              PC                Routine            Line 
>> Source
>> cpmd-50k.x         00000000004275EA  Unknown               Unknown
>> Unknown
>> cpmd-50k.x         0000000000426EDF  Unknown               Unknown
>> Unknown
>> cpmd-50k.x         00000000004657A4  Unknown               Unknown
>> Unknown
>> cpmd-50k.x         000000000040B964  Unknown               Unknown
>> Unknown
>> cpmd-50k.x         000000000040B3F5  Unknown               Unknown
>> Unknown
>> cpmd-50k.x         000000000040B3AA  Unknown               Unknown
>> Unknown
>> libc.so.6          0000003B6B91C3FB  Unknown               Unknown
>> Unknown
>> cpmd-50k.x         000000000040B2EA  Unknown               Unknown
>> Unknown
>>
>> The output file has three lines
>>
>> FREEM | COULD NOT FIND POINTER           182934130704
>>
>>  ==16:40===========================================================
>>  NAME       FIRST ADDRESS          LAST ADDRESS              LENGTH
>>
>> When I try to compile in the debug mode (with "-g") the process stops 
>> with
>> an error message
>>
>> /lib/cpp -P -C -traditional -D__Linux -D__PGI -DFFT_DEFAULT -DPARALLEL
>> -DMYRINET -DLINUX_IFC -DINTEL_MKL   ./formf.F ./formf.f
>> mpif77 -g -c -r8 -w95 -O2 -pc64 -tpp7 -unroll -cm -tune pn4 -arch pn4
>> /formf.f -o  ./formf.o
>> fortcom: Severe: **Internal compiler error: segmentation violation signal
>> raised** Please report this error along with the circumstances in which 
>> it
>> occurred in a Software Problem Report.  Note: File and line given may not
>> be explicit cause of this error.
>>
>> compilation aborted for ./formf.f (code 3)
>> make: *** [formf.o] Error 3
>>
>> I have made a try to increase the stack size (inserting 'ulimit -s
>> unlimited' in the .profile). However the result is the same.
>>
>> As I understand it seems to be an ICF problem. However this 9.1 version
>> works fine on other platforms (not em64t).
>>
>> What could be done to solve the issue?
>>
>> Vladimir
>>
>>
>>
>>
>> _______________________________________________
>> CPMD-list mailing list
>> CPMD-list at cpmd.org
>> http://cpmd.org/mailman/listinfo/cpmd-list
>
> -- 
> ===========================================================================
> Nisanth N. Nair
> Lehrstuhl fuer Theoretische Chemie   Tel: ++49 (0)234 32 22121
> Ruhr-Universitaet Bochum                     Fax: ++49 (0)234 32 14045
> Universitaetstrasse 150
> D-44801 Bochum, Germany
> Email: nisanth.nair at theochem.ruhr-uni-bochum.de
> Web  : http://www.theochem.ruhr-uni-bochum.de
> ===========================================================================
>
> 



More information about the CPMD-list mailing list