[Wien] compilation problem IBM p5 AIX5.3

Fri Sep 15 13:45:06 CEST 2006

Dear Torsten and Roberto  
Thanks for your answers,  but I tried the compilation and I have obtained errors again .  
This is what we did in order to obtain the smallest amount of errors:  
a) make the liblapack and libblas libraries by hand, with the following options:  
#------------------------------------------------------------------------------  
FC = xlf90  
FOPT     = -q64 -O3 -qcache=auto -qhot -qtune=auto -qarch=power5 -qstrict  
FGEN     =  
DESTDIR  = ../liblapack_lapw.a (Similar options were used for making libblas)  
#------------------------------------------------------------------------------  
b) After this we need to execute by hand:  
ar -rv -X32_64 ../liblapack_lapw.a lapack_lapw1.o zhpev.o zhpevx.o zladiv.o zlazro.o dlap  
y2.o dlapy3.o, where it is necessary to include the -X32_64   
The problem seems to be related with 32 bit versus 64 bit *.o files that need to be included in a common library. Am I right ?  
After this almost all the main programs were successfully compiled.
The final output-file after the compilation is attached to this email.
For example, in the case of SRC_aim, the compile.msg is:
1586-119 (U) The 32-bit file "d1mach.o" is being linked in 64-bit mode, or vice versa.
make: 1254-004 The error code from the last command is 1.
Once again the problem is related with 32bit vs 64bit generated files.
Any advice?
Thank you very much (again!)
Regards
Ricardo

-------------------------------------------------------------------------
-----   Ing. Quím. Ricardo Faccio

Mail: Cryssmat-Lab., Cátedra de Física, DETEMA
Facultad de Química, Universidad de la República
     Av. Gral. Flores 2124, C.C. 1157
     C.P. 11800, Montevideo, Uruguay.
E-mail: rfaccio at fq.edu.uy
Phone: 598 2 9241860 Int. 109
           598 2 9290705
Fax:    598 2 9241906
Web:  http://cryssmat.fq.edu.uy/ricardo/ricardo.htm
----- Original Message ----- 
From: "Roberto Iglesias" <roberto.iglesias at psi.ch>
To: "A Mailing list for WIEN2k users" <wien at zeus.theochem.tuwien.ac.at>
Sent: Thursday, September 14, 2006 3:49 AM
Subject: Re: [Wien] compilation problem IBM p5 AIX5.3

> Hello Ricardo
> 
>  To add on what Torsten wrote, we had some problems with compiling in an IBMSP4 machine and the responsible 
> finally used these settings:
> 
> current:FOPT:-q64 -O5  -qarch=pwr4 -qtune=pwr4 -qcache=auto -qstrict
> current:FPOPT:-q64 -O5 -qarch=pwr4 -qtune=pwr4 -qcache=auto -qstrict
> current:LDFLAGS:-q64 -O5 -qipa=missing=isolated -L../SRC_lib
> current:DPARALLEL:'-WF,-DParallel'
> current:R_LIBS:-lessl -llapack_lapw
> current:RP_LIBS:-L/apps/scalapack/lib -lscalapack_pwr4_64
> /apps/scalapack/lib/blacs_MPI-pwr4_64-0.a
> /apps/scalapack/lib/blacsF77init_MPI-pwr4_64-0.a
> 
> Hope it helps
> 
> Roberto
> 
> 
> ------------------------------------------
> Roberto Iglesias
> High Temperature Materials Project
> Laboratory for Materials Behaviour
> Nuclear Energy and Safety Department
> OHLD/013
> PAUL SCHERRER INSTITUT
> CH-5232 Villigen PSI
> phone: +41 (0)56 310 54 81
> fax:   +41 (0)56 310 35 65
> e-mail: roberto.iglesias at psi.ch
> Internet: www.psi.ch
> -----------------------------------------
> 
> 
> 
> Torsten Andersen wrote:
>> Hello Ricardo,
>> 
>> try this in the lapack_lapw Makefile:
>> 
>> FOPT = -q64 -O3 -qcache=auto -qhot -qtune=auto -qarch=power5 -qstrict
>> 
>> I think that the -qstrict is the most important option here.
>> 
>> You can safely remove the -bmaxdata option, since it is useless anyway 
>> in 64-bit mode.
>> 
>> Best regards,
>> Torsten Andersen.
>> 
>> 
>> Ricardo Faccio wrote:
>>> Dear Wien users:
>>> I am trying to install Wien2k04_06 in a IBM power5, AIX 5.3 with the 
>>> following software:
>>> -ESSL4
>>> -XLF90
>>> -XLC
>>> I follow the instructions on the web 
>>> (http://www.wien2k.at/reg_user/faq/ibm.html) including some  
>>> recomendations of siteconfig_lapw.
>>> The settings were:
>>> # Compiler options are: -q64 -O5 -qarch=power5
>>> # Linker options are: -q64 -O5 -qipa=missing=isolated -L../SRC_lib 
>>> -bmaxdata:0x80000000
>>> # Flags: -lessl -llapack
>>> After this I had too many errors
>>> Then I tried to edit the makefile of the lapack library separatly in 
>>> order to be compiled with the following settings:
>>>
>>> ###############################################################################
>>>
>>> FC= xlf90
>>>
>>> FOPT = -q64 -O5 -qarch=power5 /  FOPT = -O5 -qarch=power5 / FOPT = -O3 
>>> -qarch=power5
>>>
>>> GEN =
>>>
>>> DESTDIR = ../
>>>
>>> LIBNAME = liblapack_lapw.a
>>>
>>> ###############################################################################
>>>
>>> and I had to many errors again. I guess the problem is related with the 
>>> lapack library (that can't be compiled in -q64 and can be 
>>> ¿¿succesfully?? compiled with the -q32 option); beacause after this all 
>>> the routines such as lapw1, lapw2, etc that uses lapack could not be 
>>> compiled.
>>> Any advice?
>>> Thanks in advance
>>> Ricardo
>>> -------------------------------------------------------------------------
>>> -----   Ing. Quím. Ricardo Faccio
>>>  
>>> Mail: Cryssmat-Lab., Cátedra de Física, DETEMA
>>> Facultad de Química, Universidad de la República
>>>      Av. Gral. Flores 2124, C.C. 1157
>>>      C.P. 11800, Montevideo, Uruguay.
>>> E-mail: rfaccio at fq.edu.uy <mailto:rfaccio at fq.edu.uy>
>>> Phone: 598 2 9241860 Int. 109
>>>            598 2 9290705
>>> Fax:    598 2 9241906
>>> Web:  http://cryssmat.fq.edu.uy/ricardo/ricardo.htm
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>> _______________________________________________
>>> Wien mailing list
>>> Wien at zeus.theochem.tuwien.ac.at
>>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>> 
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20060915/6d2c0339/attachment.html
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: salida14-9-06.txt
Url: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20060915/6d2c0339/salida14-9-06.txt