mixed program c/fortran

Hi all, I am a beginner in programming in C, C++ as well as fortran.

I am trying to call a fortran subroutine from a c program. I found documentation on how to do that when the arguments of the subroutine are classical, such as integer...but how do you pass a fortran subroutine that has itself a function or another subroutine as an argument to a c program ?

Here is the c code

Code:

---



#include <stdio.h>
#include <stdlib.h>
#include <math.h>
#include "f77.h"


extern "C"{
void LMQN_(int *IERROR, int *N, double X[], double *F, double G[], double W[], int *LW, double *SFUN_, int *MSGLVL
, int *MAXIT, int *MAXFUN, double *ETA, int *STEPMX, double *ACCRCY, double *XTOL);
}


int
main(void)
{
int IERROR,i;
int N=200,LW=4000,MSGLVL,MAXIT,MAXFUN,STEPMX=10;
double F=-1000.0,SFUN_;
double G[300],W[4000],X[300],tipo[300];
double ETA=0.25,ACCRCY=1.0e-15,XTOL;
FILE *inputFPtr;

XTOL=sqrt(ACCRCY);
MAXIT=N/2;
MAXFUN=150*N;


  if((inputFPtr=fopen("conf1_deform.dat","r"))==NULL){
    printf("can't read conf1_deform.dat.dat\n");
    exit(1);
  }


  for(i=0;i<100;i++) {
    fscanf(inputFPtr,"%lf%lf%lf",&tipo[i],&X[i],&X[i+100]);
  }


LMQN_(&IERROR, &N, X, &F, G, W, &LW, SFUN_, &MSGLVL, &MAXIT, &MAXFUN, &ETA, &STEPMX, &ACCRCY, &XTOL);
             
}

---

Thanks all

Mike

You'd have to read the instructions which come with your compiler on how to create mixed-language programs.

Hi all again, thanks for the answer but has anyone a bit more information to give me on this topic ??

Do you mean to say that you want to pass a function pointer to a C function into a fortran function? Or are you passing a fortran function to a C function, or something else?

[I have no idea as to the answer to the question, I'm just attempting to clarify what it is that you want answered, as that would help someone who actually knows the answer].

And I'd also like to make the note that combining languages is always dependant on the compilers involved, so it would be good if you told us what compiler suite(s) you are using.

--
Mats

Hi,

I will try and clarify.

What I want to do is to include an optimization code from netlib called TN/TNBC in a c code.

The fortran code is working well, the main code is the following.

Code:

---

an C***********************************************************************
C CUSTOMIZED, NO BOUNDS
C***********************************************************************
C MAIN PROGRAM TO MINIMIZE A FUNCTION (REPRESENTED BY THE ROUTINE SFUN)
C OF N VARIABLES X - CUSTOMIZED VERSION
C
      DOUBLE PRECISION  X(300), F, G(300), W(4000)
      DOUBLE PRECISION  ETA, ACCRCY, XTOL, STEPMX, DSQRT
      EXTERNAL          SFUN
        DOUBLE PRECISION dummy
        integer i
C
C SET UP FUNCTION AND VARIABLE INFORMATION
C N  - NUMBER OF VARIABLES
C X  - INITIAL ESTIMATE OF THE SOLUTION
C F  - ROUGH ESTIMATE OF FUNCTION VALUE AT SOLUTION
C LW - DECLARED LENGTH OF THE ARRAY W
C


        open(33,file="conf1_deform.dat",status='unknown')
        do i=1,100
        read(33,*) dummy,X(i),X(i+100)
        enddo
        close(33)


      N  = 200
!      DO 10 I = 1,N
!        X(I) = I / FLOAT(N+1)
!10    CONTINUE
      F  = 1.D0
      LW = 4000
C
C SET UP CUSTOMIZING PARAMETERS
C ETA    - SEVERITY OF THE LINESEARCH
C MAXFUN - MAXIMUM ALLOWABLE NUMBER OF FUNCTION EVALUATIONS
C XTOL  - DESIRED ACCURACY FOR THE SOLUTION X*
C STEPMX - MAXIMUM ALLOWABLE STEP IN THE LINESEARCH
C ACCRCY - ACCURACY OF COMPUTED FUNCTION VALUES
C MSGLVL - DETERMINES QUANTITY OF PRINTED OUTPUT
C          0 = NONE, 1 = ONE LINE PER MAJOR ITERATION.
C MAXIT  - MAXIMUM NUMBER OF INNER ITERATIONS PER STEP
C
      MAXIT  = N/2
      MAXFUN = 150*N
      ETA    = .25D0
      STEPMX = 1.D1
      ACCRCY = 1.D-15
      XTOL  = DSQRT(ACCRCY)
      MSGLVL = 1
C
C MINIMIZE THE FUNCTION
C
      CALL LMQN (IERROR, N, X, F, G, W, LW, SFUN,
    *    MSGLVL, MAXIT, MAXFUN, ETA, STEPMX, ACCRCY, XTOL)
C
C PRINT THE RESULTS
C
      IF (IERROR .NE. 0) WRITE(*,800) IERROR
      IF (MSGLVL .GE. 1) WRITE(*,810)
      IF (MSGLVL .GE. 1) WRITE(*,820) (I,X(I),I=1,N)
      STOP
800  FORMAT(//,' ERROR CODE =', I3,/)
810  FORMAT(10X, 'CURRENT SOLUTION IS ',/14X, 'I', 11X, 'X(I)')
820  FORMAT(10X, I5, 2X, 1PD22.15)
      END
C
C

      SUBROUTINE SFUN (N, X, F, G)
      DOUBLE PRECISION  X(N), G(N), F, T
        integer i,j
          DOUBLE PRECISION  Vp,v,cutoff,pot_cutoff,dx,dy,r2
        DOUBLE PRECISION  sigma_xy,Fx,Fy,r(100)


        open(33,file="conf1_deform.dat",status='unknown')
        do i=1,100
        read(33,*) r(i)
        enddo
        close(33)

        v=0.d0
        F = 0.D0
        do i=1,100
        G(i) = 0.d0
        G(i+100) = 0.d0
        enddo

        do i=1,100
        do j=1,100
        if (i.ne.j) then
        sigma_xy=(r(i)+r(j))/2.d0
        cutoff=2.d0**(7.d0/6.d0)*sigma_xy
        pot_cutoff=4*((sigma_xy*sigma_xy/(cutoff*cutoff))**6.d0-
    $        (sigma_xy*sigma_xy/(cutoff*cutoff))**3.d0)
        dx=X(i)-X(j)
        dy=X(i+100)-X(j+100)
        if (dx.ge.10.3975/2.d0) dx=dx-10.3975
        if (dx.le.-10.3975/2.d0) dx=dx+10.3975
        if (dy.ge.10.3975/2.d0) dy=dy-10.3975
        if (dy.le.-10.3975/2.d0) dy=dy+10.3975

        r2=dx*dx+dy*dy

        if (r2.le.cutoff*cutoff) then
        Vp=4*((sigma_xy*sigma_xy/r2)**6-(sigma_xy*sigma_xy/r2)**3)-pot_cutoff
!        print *, i, j, Vp
        F=F+Vp/2.0

        Fx=48.d0*sigma_xy**12.d0*dx/r2**7.d0-
    $    24.d0*sigma_xy**6.d0*dx/r2**4.d0
        Fy=48.d0*sigma_xy**12.d0*dy/r2**7.d0-
    $    24.d0*sigma_xy**6.d0*dy/r2**4.d0
        G(i) =  G(i) - Fx
        G(i+100) = G(i+100) - Fy         
        endif
       
        endif
          enddo
        enddo       



C
C ROUTINE TO EVALUATE FUNCTION (F) AND GRADIENT (G) OF THE OBJECTIVE
C FUNCTION AT THE POINT X
C

!      F = 0.D0
!      DO 10 I = 1,N
!        T    = 2*X(I) - I
!        F    = F + T*T
!        G(I) = 2.D0 * T
!10    CONTINUE
 
!    RETURN

      END

---

The previous code I posted a c code is aimed at doing the same thing. This is just a test later I will call the subroutine LMQN from another c code that I am using. My difficulty is that I DON'T KNOW HOW TO CALL A SUBROUTINE FROM C (LMQN) THAT HAS AS ARGUMENT ANOTHER SUBROUTINE (SFUN).

Thank you for your help.
If there are references on this, please let me know.

To compile my c codes I use gcc or g++.

Mike

Ok, and are you using g77 to compile your fortran?

Have you tried declaring SFUN() in your C code:

Code:

---

extern "C" void  SFUN _(int *N, double *X, double *F, double *G);

---

[I'm not 100% sure that's the right format, but something along those lines].

--
Mats

Hi I really appreciate your help.

I use this syntax

Code:

---

extern "C"{
void SFUN_(int *N, double X[], double *F, double G[]);
}

---

I compile in this way

g++ -c minim.c
ifort -c tn.f blas.f sfun.f
g++ minim.o tn.o blas.o sfun.o

and I get this error message (I cut it as it goes on and on)

/usr/bin/ld: AVERTISSEMENT: architecture i386 du fichier d'entrée « tn.o » est incompatible avec la sortie i386:x86-64
/usr/bin/ld: AVERTISSEMENT: architecture i386 du fichier d'entrée « blas.o » est incompatible avec la sortie i386:x86-64
/usr/bin/ld: AVERTISSEMENT: architecture i386 du fichier d'entrée « sfun.o » est incompatible avec la sortie i386:x86-64
minim.o(.text+0x1d6): In function `main':
: undefined reference to `LMQN_'
tn.o(.text+0x136): In function `tn_':
: undefined reference to `for_write_seq_fmt'
tn.o(.text+0x16d): In function `tn_':
: undefined reference to `for_write_seq_fmt'
tn.o(.text+0x1a1): In function `tn_':
: undefined reference to `for_write_seq_fmt'
tn.o(.text+0x1d7): In function `tn_':
: undefined reference to `for_write_seq_fmt'
tn.o(.text+0x204): In function `tn_':
: undefined reference to `for_write_seq_fmt_xmit'


My problem is how to pass the subroutine SFUN_ as an argument inside the other subroutine LMQN_.

For the moment I passing like that

Code:

---

extern "C"{
void LMQN_(int *IERROR, int *N, double X[], double *F, double G[], double W[], int *LW, void *SFUN_, int *MSGLVL, int *MAXIT, int *MAXFUN, double *ETA, int *STEPMX, double *ACCRCY, double *XTOL);
}

---

and

Code:

---

LMQN_(&IERROR, &N, X, &F, G, W, &LW, &SFUN_, &MSGLVL, &MAXIT, &MAXFUN, &ETA, &STEPMX, &ACCRCY, &XTOL);

---

Thanks for the suggestion any other ideas ???

First of all, "architecture i386 du fichier d'entrée « tn.o » est incompatible avec la sortie i386:x86-64" says that your file contains x86_32 functions, so you should pass -m32 to g++ to match the functions to your C++ code.

This would be a more correct version of passing SFUN_ to your function.

Code:

---

void LMQN_(int *IERROR, int *N, double X[], double *F, double G[], double W[], int *LW, void (*SFUN_)(int *N, double X[], double *F, double G[]), int *MSGLVL, int *MAXIT, int *MAXFUN, double *ETA, int *STEPMX, double *ACCRCY, double *XTOL);

---

--
Mats

This makes more sence.

Here is what I get now.

g++ -c minim.c -m32

minim.c: In function `int main()':
minim.c:45: erreur: ne peut convertir « double » à « void (*)(int*, double*, double*, double*) » pour l'argument « 8 » vers « void LMQN_(int*, int*, double*, double*, double*, double*, int*, void (*)(int*, double*, double*, double*), int*, int*, int*, double*, int*, double*, double*) »

(sorry for all this french)

You didn't mention how I should pass the subroutine SFUN_ as an argument when I call LMQN_ in the c code. For the moment I am doing

Code:

---

LMQN_(&IERROR, &N, X, &F, G, W, &LW, SFUN_, &MSGLVL, &MAXIT, &MAXFUN, &ETA, &STEPMX, &ACCRCY, &XTOL);

---

I also tried &SFUN_ and it gives the above error.
What is the proper syntax ?

By the way do you know any usefull reference on these topics?

Thanks a lot.

Mike

Trying to read the french and translating to English, I get the impression that it says that it can't translate a double to void (*)(...) for argument 8 - which seems strange to me.

No, I don't know of any references for this.

--
Mats

ANY SPECIALIST ON HOW TO MIX THE LANGUAGES FORTRAN AND C see threads above???

Thanks I am stuck there for 5 days and I suspect it is trivial.

Thanks everyone and Mats especially.

Quote:

---

Originally Posted by mimi1981

ANY SPECIALIST ON HOW TO MIX THE LANGUAGES FORTRAN AND C see threads above???

Thanks I am stuck there for 5 days and I suspect it is trivial.

Thanks everyone and Mats especially.

---

Just to clarify, the current problem is a pure C (or C++) problem, not a Fortran mix with C++ problem - or at least, the compiler error should be solved with fixing the C++ code, it may not be completely correct in other ways, and that may cause problems in the interface, but at the moment, it's a pure C++ problem.

Can you post the whole C++ code as it stands now?

--
Mats

Here is the code (with my question in red).

Code:

---

#include <stdio.h>
#include <stdlib.h>
#include <math.h>
#include "f77.h"    [COLOR="I don't know what this does ?"][/COLOR]


extern "C"{
void SFUN_(int *N, double X[], double *F, double G[]);
}


//extern "C" void  SFUN_(int *N, double *X, double *F, double *G);

extern "C"{
void LMQN_(int *IERROR, int *N, double X[], double *F, double G[], double W[], int *LW, void (*SFUN_)(int *N, doub
le X[], double *F, double G[]), int *MSGLVL, int *MAXIT, int *MAXFUN, double *ETA, int *STEPMX, double *ACCRCY, do
uble *XTOL);
}


int
main(void)
{
int IERROR,i;
int N=200,LW=4000,MSGLVL,MAXIT,MAXFUN,STEPMX=10;
double F=-1000.0,SFUN_;
double G[300],W[4000],X[300],tipo[300];
double ETA=0.25,ACCRCY=1.0e-15,XTOL;
FILE *inputFPtr;

XTOL=sqrt(ACCRCY);
MAXIT=N/2;
MAXFUN=150*N;


  if((inputFPtr=fopen("conf1_deform.dat","r"))==NULL){
    printf("can't read conf1_deform.dat.dat\n");
    exit(1);
  }


  for(i=0;i<100;i++) {
    fscanf(inputFPtr,"%lf%lf%lf",&tipo[i],&X[i],&X[i+100]); [COLOR="How could I pass tipo[i] to the function SFUN without changing its signature ie the number of its arguments ???"][/COLOR]
  }


LMQN_(&IERROR, &N, X, &F, G, W, &LW, &SFUN_, &MSGLVL, &MAXIT, &MAXFUN, &ETA, &STEPMX, &ACCRCY, &XTOL); [COLOR="WHAT SHOULD I TYPE THERE &SFUN_, SFUN...?????"][/COLOR]



}

---

The function sfun.f is the following

Code:

---

      SUBROUTINE SFUN (N, X, F, G)
      DOUBLE PRECISION  X(N), G(N), F, T
        integer i,j
        DOUBLE PRECISION  Vp,v,cutoff,pot_cutoff,dx,dy,r2
        DOUBLE PRECISION  sigma_xy,Fx,Fy,r(100)


        open(33,file="conf1_deform.dat",status='unknown')
        do i=1,100
        read(33,*) r(i)
        enddo
        close(33)

        v=0.d0
        F = 0.D0
        do i=1,100
        G(i) = 0.d0
        G(i+100) = 0.d0
        enddo

        do i=1,100
        do j=1,100
        if (i.ne.j) then
        sigma_xy=(r(i)+r(j))/2.d0
        cutoff=2.d0**(7.d0/6.d0)*sigma_xy
        pot_cutoff=4*((sigma_xy*sigma_xy/(cutoff*cutoff))**6.d0-
    $        (sigma_xy*sigma_xy/(cutoff*cutoff))**3.d0)
        dx=X(i)-X(j)
        dy=X(i+100)-X(j+100)
        if (dx.ge.10.3975/2.d0) dx=dx-10.3975
        if (dx.le.-10.3975/2.d0) dx=dx+10.3975
        if (dy.ge.10.3975/2.d0) dy=dy-10.3975
        if (dy.le.-10.3975/2.d0) dy=dy+10.3975

        r2=dx*dx+dy*dy

        if (r2.le.cutoff*cutoff) then
        Vp=4*((sigma_xy*sigma_xy/r2)**6-(sigma_xy*sigma_xy/r2)**3)-pot_cutoff
!        print *, i, j, Vp
        F=F+Vp/2.0

        Fx=48.d0*sigma_xy**12.d0*dx/r2**7.d0-
    $    24.d0*sigma_xy**6.d0*dx/r2**4.d0
        Fy=48.d0*sigma_xy**12.d0*dy/r2**7.d0-
    $    24.d0*sigma_xy**6.d0*dy/r2**4.d0
        G(i) =  G(i) - Fx
        G(i+100) = G(i+100) - Fy
        endif

        endif
        enddo
        enddo


      END

---

and F77.h is (I found that on the internet maybe it does not make sense to put it as an include ?????

Code:

---

#ifndef Lib_F77_h
#define Lib_F77_h

#ifdef WIN32
#define Lib_stdcall __stdcall
#else
#define Lib_stdcall
#endif

#endif

---

THANKS A LOT

It was probably in your original code, but changing this line:

Code:

---

double F=-1000.0,SFUN_;

---

into

Code:

---

double F=-1000.0;

---

makes the code compile with g++ on my machine. It doesn't link because I don't have the libraries. And there may of course be other problems.

As to how you pass "tipo[i]", I think you would not do that, since sfun() reads that from the datafile in itself.

If you want to read the file only once, you would have to change the FORTRAN code around to take another argument double tipo[] in both LMQN and SFUN, and pass tipo from LMQN to SFUN.

--
Mats

So that is one stupid mistake sorted. Thanks and sorry about that.

One question you haven't answered is how to pass SFUN as an argument to the function LMPQ when I call it through

Code:

---

LMQN_(&IERROR, &N, X, &F, G, W, &LW, &SFUN_, &MSGLVL, &MAXIT, &MAXFUN, &ETA, &STEPMX, &ACCRCY, &XTOL);

---

Is this the right syntax ?? or is it something else SFUN_,SFUN ....??

If this is correct, I think we are back to the initial questions about the way to link C++ and fortran functions.

I do

g++ -c minim.c -m32
ifort -c tn.f
ifort -c blas.f
ifort -c sfun.f
g++ minim.o blas.o sfun.o tn.o -m32

and I get the horrible
/usr/bin/ld: AVERTISSEMENT: architecture i386 du fichier d'entrée « minim.o » est incompatible avec la sortie i386:x86-64
/usr/bin/ld: AVERTISSEMENT: architecture i386 du fichier d'entrée « tn.o » est incompatible avec la sortie i386:x86-64
/usr/bin/ld: AVERTISSEMENT: architecture i386 du fichier d'entrée « sfun.o » est incompatible avec la sortie i386:x86-64
minim.o(.text+0x169): In function `main':
: undefined reference to `SFUN_'
minim.o(.text+0x1a7): In function `main':
: undefined reference to `LMQN_'
tn.o(.text+0x136): In function `tn_':
: undefined reference to `for_write_seq_fmt'
tn.o(.text+0x16d): In function `tn_':
: undefined reference to `for_write_seq_fmt'
tn.o(.text+0x1a1): In function `tn_':
: undefined reference to `for_write_seq_fmt'
tn.o(.text+0x1d7): In function `tn_':
: undefined reference to `for_write_seq_fmt'
tn.o(.text+0x204): In function `tn_':
: undefined reference to `for_write_seq_fmt_xmit'
tn.o(.text+0x226): In function `tn_':
: undefined reference to `for_write_seq_fmt_xmit'
tn.o(.text+0x241): In function `tn_':
: undefined reference to `for_write_seq_fmt_xmit'
tn.o(.text+0x2bd): In function `lmqn_.':

and on and on....

Can you do "file blas.o" and post the output?

Maybe it's ifort than need as -m32 on it?

--
Mats

file blas.o gives

blas.o: ELF 32-bit LSB relocatable, Intel 80386, version 1 (GNU/Linux), not stripped


Can you answer to my last question concerning how to pass the function as an argument also ??

A function name will expand to the address of the function automagically, so SFUN_ and &SFUN_ is the same thing.

So, blas.o is a 32-bit object file. You need to use -m32 on the g++ compiler, and you do show that, so I'm not sure what is wrong - perhaps you need to have it BEFORE the file itself (e.g. "g++ -m32 x.cpp") - check with "file x.o" when you have run g++ on your source file.

Edit: I think the write_file_seq or whatever it is called is a fortran function, so you probably need to add some library to your linking list as well. Whilst looking for the correct name of that library, I found this: http://www.chiralcomp.com/support/mi...p/linking.html


Edit2: And this: http://softwarecommunity.intel.com/i.../30246677.aspx
Edit3: And: http://softwarecommunity.intel.com/i...ad/223602.aspx

--
Mats

So I have tested what they recommend and here it is. I use two simple test programs called test.f and test.c

1) ifort

ifort -v test.f gives:

Version 9.0
/opt/intel/fc/9.0/bin/fortcom -D__INTEL_COMPILER=900 -D__ELF__ -D__INTEL_COMPILER_BUILD_DATE=20051201 -D__unix__ -D__unix -D__linux__ -D__linux -D__gnu_linux__ -Dunix -Dlinux -D__i386 -D__i386__ -Di386 -mGLOB_pack_sort_init_list -I. -I/opt/intel/fc/9.0/include -I/opt/intel/fc/9.0/substitute_headers -I/usr/include -O2 -mP1OPT_version=900 -mGLOB_source_language=GLOB_SOURCE_LANGUAGE_F90 -mGLOB_tune_for_fort -mGLOB_use_fort_dope_vector -mP2OPT_static_promotion -mP1OPT_print_version=FALSE -mP3OPT_use_mspp_call_convention -mCG_use_gas_got_workaround=F -mP2OPT_align_option_used=TRUE -mGLOB_options_string=-v -mGLOB_cxx_limited_range=FALSE -mGLOB_as_output_backup_file_name=/tmp/ifortZBIJSOas_ -mGLOB_machine_model=GLOB_MACHINE_MODEL_IA32_NONE -mP2OPT_subs_out_of_bound=FALSE -mIPOPT_ninl_user_level=2 -mIPOPT_activate -mIPOPT_lite -mP2OPT_hlo_level=2 -mP2OPT_hlo -mIPOPT_obj_output_file_name=/tmp/ifortSzlKyt.o "-mGLOB_linker_version=2.15.92.0.2 20040927" -mP3OPT_asm_target=P3OPT_ASM_TARGET_GAS -mGLOB_obj_output_file=/tmp/ifortSzlKyt.o -mGLOB_source_dialect=GLOB_SOURCE_DIALECT_FORTRAN -mP1OPT_source_file_name=test.f test.f
ld /usr/lib/crt1.o /usr/lib/crti.o /usr/lib/gcc/x86_64-redhat-linux/3.4.6/32/crtbegin.o -dynamic-linker /lib/ld-linux.so.2 -m elf_i386 -o a.out /opt/intel/fc/9.0/lib/for_main.o /tmp/ifortSzlKyt.o -L/opt/intel/fc/9.0/lib -L/usr/lib -Bstatic -lifport -lifcore -Bdynamic -limf -lm -Bstatic -lirc -Bdynamic -lc -Bstatic -lirc_s -Bdynamic -ldl -lc /usr/lib/gcc/x86_64-redhat-linux/3.4.6/32/crtend.o /usr/lib/crtn.o
rm /tmp/ifortjmtCd9gas

rm /tmp/ifortZBIJSOas_

rm /tmp/ifortEEwSxuld

rm /tmp/ifortucUkdaarg

rm /tmp/ifortSzlKyt.o


2) g++

g++ test.c gives

Lecture des spécification à partir de /usr/lib/gcc/x86_64-redhat-linux/3.4.6/specs
Configuré avec: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --disable-checking --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-java-awt=gtk --host=x86_64-redhat-linux
ModÚle de thread: posix
version gcc 3.4.6 20060404 (Red Hat 3.4.6-9)
/usr/libexec/gcc/x86_64-redhat-linux/3.4.6/cc1plus -quiet -v -D_GNU_SOURCE test.c -quiet -dumpbase test.c -mtune=k8 -auxbase test -version -o /tmp/ccucfvZc.s
le répertoire « /usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../x86_64-redhat-linux/include » est ignoré car inexistant
la recherche pour #include "..." débute ici :
la recherche pour #include <...> débute ici:
/usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../include/c++/3.4.6
/usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../include/c++/3.4.6/x86_64-redhat-linux
/usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../include/c++/3.4.6/backward
/usr/local/include
/usr/lib/gcc/x86_64-redhat-linux/3.4.6/include
/usr/include
Fin de la liste de recherche.
GNU C++ version 3.4.6 20060404 (Red Hat 3.4.6-9) (x86_64-redhat-linux)
compiled by GNU C version 3.4.6 20060404 (Red Hat 3.4.6-9).
heuristiques GGC: --param ggc-min-expand=100 --param ggc-min-heapsize=131072
as -V -Qy -o /tmp/ccYbsOyj.o /tmp/ccucfvZc.s
Version de l'assembleur GNU 2.15.92.0.2 (x86_64-redhat-linux) utilisant la version BFD 2.15.92.0.2 20040927
/usr/libexec/gcc/x86_64-redhat-linux/3.4.6/collect2 --eh-frame-hdr -m elf_x86_64 -dynamic-linker /lib64/ld-linux-x86-64.so.2 /usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../lib64/crt1.o /usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../lib64/crti.o /usr/lib/gcc/x86_64-redhat-linux/3.4.6/crtbegin.o -L/usr/lib/gcc/x86_64-redhat-linux/3.4.6 -L/usr/lib/gcc/x86_64-redhat-linux/3.4.6 -L/usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../lib64 -L/usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../.. -L/lib/../lib64 -L/usr/lib/../lib64 /tmp/ccYbsOyj.o -lstdc++ -lm -lgcc_s -lgcc -lc -lgcc_s -lgcc /usr/lib/gcc/x86_64-redhat-linux/3.4.6/crtend.o /usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../lib64/crtn.o


I am still reading if the next seems obvious to you, please let me know.

Thanks a lot.
Mike

I believe you want to add:

Code:

---

-L/opt/intel/fc/9.0/lib -Bstatic -lifport -lifcore -Bdynamic -limf

---

to the line:

Code:

---

g++ minim.o blas.o sfun.o tn.o -m32

---

I'm by far not convinced that is enough to make it work, but it's unlikely to make anything worse [since adding more libraries doesn't make any difference if the library isn't being used, but not having a library that is used does definitely cause problems].

--
Mats

Thanks, but the problem is that when I do

g++ minim.o tn.o sfun.o blas.o -L/opt/intel/fc/9.0/lib -Bstatic -lifport -lifc

I get

/usr/bin/ld: escamotage incompatible /opt/intel/fc/9.0/lib/libifport.so.5 lors de la recherche de libifport.so.5
/usr/bin/ld: ne peut trouver libifport.so.5
collect2: ld a retourné 1 code d'état d'exécution

And when I do

g++ minim.o tn.o sfun.o blas.o -m32 -L/opt/intel/fc/9.0/lib -Bstatic -lifport -lifc

I get

minim.o(.text+0x169): In function `main':
: undefined reference to `SFUN_'
minim.o(.text+0x1a7): In function `main':
: undefined reference to `LMQN_'
collect2: ld a retourné 1 code d'état d'exécution



By the way If you or anyone else wants to have a go and try to call this library from a c++ program here is the link for the Netlib truncated newton optimization method:

http://iris.gmu.edu/~snash/nash/software/tn.f

Thanks a lot.

Mike

It would help cut down on the bandwidth if you could write a 10-line Fortran program that you want to link with a 10-line C program.

Something simple, like add two numbers together.

At least then we can all see exactly what's going on.

You may figure it out yourself from only having one or two error messages to deal with instead of dozens.

Also, since a lot of members here don't speak/read french, translating the error messages to English would increase the number of people that may know how to fix the problem (it does not have to be translated precisely, but a rough translation is nearly always enough) [I can just about figure out what is fish or meat when I see a french menu, and I couldn't ask for the directions to the toilet - but sometimes I can sort of guess the meaning of an error message because the words are similar enough to words that I know from English or Swedish].

Salem's suggestion is also very good - make a simple test-case first, then try more complex stuff stepwise - then you will know which change made the difference.

I'm fairly convinced you need the -m32 option on your g++ line, even though I don't actually understand what the error message says for your first link attempt.

In your -m32 link attemp, it seems like the name of the functions are wrong - perhaps you can use "nm sfun.o" to see what the actual names are.



--
Mats

Just one (offtopic) question. Would (on linux) "LOCALE=en_US gcc ..." make it in english? (I have only one locale (english) installed and can't test).

EDIT: I mean, is english hardcoded into gcc so it doesn't need locale files for english?

Hi everyone.

Salem I followed your advice and tested my problem on very simple programs.

First the c code test2.c that calls a fortran function

Code:

---

#include <stdio.h>
#include <stdlib.h>
#include <math.h>

void SFUN(int *N, double *X, double *F, double *G)
{
int i;
double T;
*F=0.0;
for (i = 0 ; i < *N ; i++)
    {
    T = 2*X[i] - i;
    *F = *F +T*T;
    G[i] = 2.0 * T;
    }
}

int main(void)
{
int N=200;
double F=-1000.0;
double G[300],X[300];

extern void pseudoLMQN_(int *N, double X[], double *F, double G[], void (*SFUN)(int *N, double X[], double *F, dou
ble G[]));


printf("F before test is %lf\n",&F);

pseudoLMQN_(&N, X, &F, G, SFUN);

printf("F after test is %lf\n",&F);

}

---

And the fortran function pseudoLMQN.f

Code:

---

      SUBROUTINE LMQN (N, X, F, G, SFUN)
      IMPLICIT          NONE
      INTEGER N
      DOUBLE PRECISION  X(N), G(N), F, FNEW
      EXTERNAL SFUN

      CALL SFUN(N,X,FNEW,G)
      F=FNEW+1.0

      RETURN
      END

---

I included the function that I pass as an argument to the fortran function in the c code. For the rest the problem is similar to the previous threads.


I compile like that :

gcc -c test2.c
ifort -c pseudoLMQN.f
gcc test2.c pseudoLMQN.o

and I get the following errors :

/usr/bin/ld: AVERTISSEMENT: architecture i386 du fichier d'entrée « pseudoLMQN.o » est incompatible avec la sortie i386:x86-64
test2.o(.text+0x100): In function `main':
: undefined reference to `pseudoLMQN_'
collect2: ld a retourné 1 code d'état d'exécution

Concerning the errors in french, it is all greek to me so I cannot really translate it. Sorry about that.

But I think there is still a problem with the way I link the two programs.


Any ideas ??

Thanks a lot.

In pseudoLMQN read

Code:

---

      SUBROUTINE pseudoLMQN(N, X, F, G, SFUN)
      IMPLICIT          NONE
      INTEGER N
      DOUBLE PRECISION  X(N), G(N), F, FNEW
      EXTERNAL SFUN

      CALL SFUN(N,X,FNEW,G)
      F=FNEW+1.0

      RETURN
      END

---

Quote:

---

Originally Posted by mimi1981

I included the function that I pass as an argument to the fortran function in the c code. For the rest the problem is similar to the previous threads.


I compile like that :

gcc -c test2.c
ifort -c pseudoLMQN.f
gcc test2.c pseudoLMQN.o

and I get the following errors :

/usr/bin/ld: AVERTISSEMENT: architecture i386 du fichier d'entrée « pseudoLMQN.o » est incompatible avec la sortie i386:x86-64
test2.o(.text+0x100): In function `main':
: undefined reference to `pseudoLMQN_'
collect2: ld a retourné 1 code d'état d'exécution

Concerning the errors in french, it is all greek to me so I cannot really translate it. Sorry about that.

But I think there is still a problem with the way I link the two programs.


Any ideas ??

Thanks a lot.

---

Right, this is the same old "you are mixing 32-bit and 64-bit code" error. You really NEED to use the -m32 when compiling AND LINKING the code you produce, as (apparently) your ifort produces 32-bit code.

--
Mats

Hi again, I added what you suggested

Now I type

gcc -c -m32 test2.c
ifort -c pseudoLMQN.f
gcc -m32 test2.c pseudoLMQN.o

and I get

test2.o(.text+0xd6): In function `main':
: undefined reference to `pseudoLMQN_'
collect2: ld a retourné 1 code d'état d'exécution


A bit better, but still ????

test2.c doesn't seem to be able to use the fortran function pseudoLMQN_ defined in pseudoLMQN.f. Any ideas anyone?

Thanks

do "nm pseudoLMQN.o |grep -i pseudoLMQN > somefile.txt" and post the content of "somefile.txt".

--
Mats

somefile.txt contains :

0000000000000000 T pseudolmqn_

??

So, you probably should not use upper-case in your filename. Fortran is not case-sensitive, so the compiler will convert all names to same case, in this case lowercase, so that the linker doesn't have to know if the language cares about case or not - it just matches the name in the object file.

--
Mats

WHAOOU

It compiled, I see what you mean I put everywhere pseudolmqn in the c code and it works !!!

Thanks, what a relief. Let's not get carried away though.

Quote:

---

Originally Posted by mimi1981

WHAOOU

It compiled, I see what you mean I put everywhere pseudolmqn in the c code and it works !!!

Thanks, what a relief. Let's not get carried away though.

---

Now, the next question is if your mixed code produces correct results!

--
Mats

Hi again

We did one little step but there are still quite a few I suspect.

Now I pass to the actual code and I do

gcc -c -m32 test1.c
ifort -c tn.f
ifort -c blas.f
gcc -m32 test1.o tn.o blas.o -lm

and I get the errors related to the linking to the libraries

tn.o(.text+0x136): In function `tn_':
: undefined reference to `for_write_seq_fmt'
tn.o(.text+0x16d): In function `tn_':
: undefined reference to `for_write_seq_fmt'
tn.o(.text+0x1a1): In function `tn_':
: undefined reference to `for_write_seq_fmt'
tn.o(.text+0x1d7): In function `tn_':
: undefined reference to `for_write_seq_fmt'
tn.o(.text+0x204): In function `tn_':
: undefined reference to `for_write_seq_fmt_xmit'
tn.o(.text+0x226): In function `tn_':
: undefined reference to `for_write_seq_fmt_xmit'
tn.o(.text+0x241): In function `tn_':
: undefined reference to `for_write_seq_fmt_xmit'
tn.o(.text+0x2bd): In function `lmqn_.':
: undefined reference to `for_write_seq_fmt'
tn.o(.text+0x48a): In function `lmqn_.':
: undefined reference to `for_write_seq_fmt'
tn.o(.text+0x4b2): In function `lmqn_.':
: undefined reference to `for_write_seq_fmt_xmit'
tn.o(.text+0x4da): In function `lmqn_.':
: undefined reference to `for_write_seq_fmt_xmit'
tn.o(.text+0x502): In function `lmqn_.':

etc....etc....


I saw on a forum someone with a similar problem got reed of these undefined... by typing -lifcore, but strangely while I can see the library associated to this command in /opt/intel/fc/9.0/lib/ it is libifcore.a the command doesn't work it gives


gcc -m32 test1.o tn.o blas.o -lm -lifcore

/usr/bin/ld: ne peut trouver -lifcore (means cannot find -lifcore ?????????)
collect2: ld a retourné 1 code d'état d'exécution


What doesn't it work ???

IDEAS ???

Thanks a lot, one day it will work !!

Try "-L/opt/intel/fc/9.0/lib/" switch to tell compiler where libraries are, specify it before -lifcore. If that won't work, maybe library is not 32-bit.

Thanks to all of you. You helped me a lot with these very annoying programming problems.

Now it is working. To summarize I compile with

ifort -c tn.f
ifort -c blas.f
gcc -m32 -g -c test1.c
gcc -g -m32 test1.o tn.o blas.o -L/opt/intel/fc/9.0/lib/ -lifcore -lg2c


and to pass in a function another fortran function as argument one as to write someting like that

Code:

---

#include <stdlib.h>
#include <math.h>

void SFUN(int *N, double *X, double *F, double *G)
{
int i;
double T;
*F=0.0;
for (i = 0 ; i < *N ; i++)
    {
    T = 2*X[i] - i;
    *F = *F +T*T;
    G[i] = 2.0 * T;
    }
}

int main(void)
{
int N=200;
double F=-1000.0;
double G[300],X[300];

extern void pseudolmqn_(int *N, double X[], double *F, double G[], void (*SFUN)(int *N, double X[], double *F, double G[]));


printf("F before test is %lf\n",&F);

pseudolmqn_(&N, X, &F, G, SFUN);

printf("F after test is %lf\n",&F);

}

---

where pseudolmqn.f is

Code:

---

      SUBROUTINE pseudolmqn(N, X, F, G, SFUN)
      IMPLICIT          NONE
      INTEGER N
      DOUBLE PRECISION  X(N), G(N), F, FNEW
      EXTERNAL SFUN

      CALL SFUN(N,X,FNEW,G)
      F=FNEW+1.0

      RETURN
      END

---

AND THIS WORKS !!!!

Hi all!

The problem is now a little bit different.

I am compiling with

gcc -m32 -g -c test1.c
ifort -c tn.f
ifort -c blas.f
gcc -g -m32 test1.o tn.o blas.o -L/opt/intel/fc/9.0/lib/ -lifcore

and it works fine

BUT now I want to replace the test code test1.c by my own code test1_KANT.c but this new code requires the library -lgsl -lgslcblas, so I try

gcc -m32 -g -c test1_KANT.c
ifort -c tn.f
ifort -c blas.f
gcc -g -m32 test1.o tn.o blas.o -L/opt/intel/fc/9.0/lib/ -lifcore -lgsl -lgslcblas

and it gives me

/usr/bin/ld: escamotage incompatible /usr/lib64/libgsl.so lors de la recherche de -lgsl
/usr/bin/ld: escamotage incompatible /usr/lib64/libgsl.a lors de la recherche de -lgsl
/usr/bin/ld: ne peut trouver -lgsl
collect2: ld a retourné 1 code d'état d'exécution

It seems that -lgsl -lgslcblas are configured to go and look in /usr/lib64/ while I suspect they should in this case look in /usr/lib/

But in
/usr/lib64/ I found

libgslcblas.a
libgslcblas.la
libgslcblas.so
libgslcblas.so.0
libgslcblas.so.0.0.0

and in /usr/lib/ only
libgslcblas.la
libgslcblas.so.0
libgslcblas.so.0.0.0
libgsl.la
libgsl.so.0
libgsl.so.0.6.0

so not the libgsl.a ???

Why can't I compile everything in 64 bits , how can I link the new code with the gsl libraries ??

PLEASE HELP
THANKS A LOT

You probably could compile everything 64-bit if you compile it with g77 - the gcc version of Fortran77. If you want to use ifort, it probably also exists in a 64-bit version.

If you can't change the compiler, you probably need to install some 32-bit "-devel" components to get the 32-bit version of those libraries.

--
Mats

Great this forum is really fantastic and Matsp in particular !!! Thanks, your advices are always spot on.

I did as you suggested

gcc -c test1_KANT.c
g77 -c tn.f
g77 -c blas.f
gcc test1_KANT.o tn.o blas.o -lgsl -lgslcblas -lg2c

-lg2c is crucial, don't ask me why.

and IT WORKS

g2c is GNU's Fortran run time library. g77, when it compilers Fortran code, converts a number of operations (aka language features) into calls of functions in the run time library. It's also possible (I haven't specifically checked) the other libraries you are using (gsl, gslcblas) also make use of the Fortran run time library. Either way, -lg2c would be required.