Problems with Fortran 90 NR

[jake_harv88]

Hey guys,

I'm having issues using some of the Fortran 90 Numerical Recipe modules. Specifically I just wanted to try use the hello_bessel program on page 936. The code for the program is as follows:

      PROGRAM hello_bessel
      USE nrtype
      USE nr, ONLY: flmoon, bessj0
      IMPLICIT NONE
      INTEGER(I4B) :: n=200,nph=2,jd
      REAL(SP) :: x,frac,ans
      call flmoon(n,nph,jd,frac)
      x=jd**0.24_sp
      ans=bessj0(x)
      write(*,*) 'Hello, Bessel: ', ans
      END PROGRAM

the nr module is very long but the flmoon and the bessj0 interfaces look like this:

      INTERFACE
              SUBROUTINE flmoon(n,nph,jd,frac)
              USE nrtype
              INTEGER(I4B), INTENT(IN) :: n,nph
              INTEGER(I4B), INTENT(OUT) :: jd
              REAL(SP), INTENT(OUT) :: frac
              END SUBROUTINE flmoon
      END INTERFACE

     INTERFACE bessj0
              FUNCTION bessj0_s(x)
              USE nrtype
              REAL(SP), INTENT(IN) :: x
              REAL(SP) :: bessj0_s
              END FUNCTION bessj0_s
!BL
              FUNCTION bessj0_v(x)
              USE nrtype
              REAL(SP), DIMENSION(:), INTENT(IN) :: x
              REAL(SP), DIMENSION(size(x)) :: bessj0_v
              END FUNCTION bessj0_v
      END INTERFACE

When I try and compile I've tried a number of different commands but I'm pretty sure the right way to do it would be to do:

ifort nrtype.f90 nr.f90 hello_bessel.f90

But whenever I do that I get:

/tmp/ifortzYtJmW.o(.text+0x2f): In function `MAIN__':
: undefined reference to `flmoon_'
/tmp/ifortzYtJmW.o(.text+0x54): In function `MAIN__':
: undefined reference to `bessj0_s_'

Any suggestions on where I'm going wrong?

[Rubberman]

I'm not familiar with ifort - is that an Intel compiler? Usually I and my physicist wife use the gnu f77 compiler. Have you tried that?

[drl]

Hi.

I created stub routines for flmoon, and bessj0_[sv] based on the interfaces. Here is a driver script that shows the high points:

#!/usr/bin/env bash

# @(#) s1	Demonstrate compile with ifort, link, stub, execute.

# Utility functions: print-as-echo, print-line-with-visual-space, debug.
# export PATH="/usr/local/bin:/usr/bin:/bin"
pe() { for _i;do printf "%s" "$_i";done; printf "\n"; }
pl() { pe;pe "-----" ;pe "$*"; }
db() { ( printf " db, ";for _i;do printf "%s" "$_i";done;printf "\n" ) >&2 ; }
db() { : ; }
C=$HOME/bin/context && [ -f $C ] && $C
version ifort

rm -f *.o *.mod a.out
pl " Source files present:"
ls *.f90

pl " Sample stub, flmoon.f90:"
cat flmoon.f90

pl " Compile, link:"
ifort nrtype.f90 nr.f90 hello_bessel.f90 bessj0_s.f90 bessj0_v.f90 flmoon.f90

pl " Results:"
./a.out

# Cleanup.
rm -f *.o *.mod a.out

exit 0

producing:

% ./s1

Environment: LC_ALL = C, LANG = C
(Versions displayed with local utility "version")
OS, ker|rel, machine: Linux, 2.6.26-2-amd64, x86_64
Distribution        : Debian GNU/Linux 5.0.8 (lenny) 
GNU bash 3.2.39
ifort (IFORT) 11.1 20090827

-----
 Source files present:
bessj0_s.f90  bessj0_v.f90  flmoon.f90	hello_bessel.f90  nr.f90  nrtype.f90

-----
 Sample stub, flmoon.f90:

	  SUBROUTINE flmoon(n,nph,jd,frac)
	  USE nrtype
	  INTEGER(I4B), INTENT(IN) :: n,nph
	  INTEGER(I4B), INTENT(OUT) :: jd
	  REAL(SP), INTENT(OUT) :: frac
	  jd =0
	  frac = 0.0
	  END SUBROUTINE flmoon


-----
 Compile, link:

-----
 Results:
 Hello, Bessel:   0.000000000000000E+000

So if your results are missing routines, how is my procedure different from yours? Perhaps that mine included the compile of flmoon, etc? ... cheers, drl