This is a follow up to my previous question. I wrapped my Fortran code in a module and now it compiles when I run:
R CMD SHLIB ./Fortran/Fpi.f90
Aparently I cannot wrap the subroutines in a module. This fortran code is working:
subroutine dboard(darts, dartsscore)
implicit none
integer, intent(in) :: darts
double precision, intent(out) :: dartsscore
double precision :: x_coord, y_coord
integer :: score, n
score = 0
do n = 1, darts
call random_number(x_coord)
call random_number(y_coord)
if ((x_coord**2 + y_coord**2) <= 1.0d0) then
score = score + 1
end if
end do
dartsscore = 4.0d0*score/darts
end subroutine dboard
subroutine pi(avepi, DARTS, ROUNDS)
implicit none
double precision, intent(out) :: avepi
integer, intent(in) :: DARTS, ROUNDS
integer :: MASTER, rank, i, n
integer, allocatable :: seed(:)
double precision :: pi_est, homepi, pirecv, pisum
interface
subroutine dboard(darts, dartsscore)
implicit none
integer, intent(in) :: darts
double precision, intent(out) :: dartsscore
end subroutine dboard
end interface
! we set it to zero in the sequential run
rank = 0
! initialize the random number generator
! we make sure the seed is different for each task
call random_seed()
call random_seed(size = n)
allocate(seed(n))
seed = 12 + rank*11
call random_seed(put=seed(1:n))
deallocate(seed)
avepi = 0
do i = 0, ROUNDS-1
call dboard(darts, pi_est)
! calculate the average value of pi over all iterations
avepi = ((avepi*i) + pi_est)/(i + 1)
end do
end subroutine pi
and this is the R code:
system("R CMD SHLIB ./Fortran/Fpi.f90")
mypi <- function(DARTS, ROUNDS) {
dyn.load("./Fortran/Fpi.so")
retvals <- .Fortran("pi", avepi = as.numeric(1), DARTS = as.integer(DARTS), ROUNDS = as.integer(ROUNDS))
return(retvals$avepi)
}
mypi(DARTS = 50000, ROUNDS = 10)
Your problem comes down to the declaration of dboard
:
double precision :: pi_est, homepi, pirecv, pisum, dboard
Here you are saying that dboard
is an external function, rather than a module procedure. This explains why there is a symbol dboard_
coming into play. You want to remove that:
double precision :: pi_est, homepi, pirecv, pisum
and instead rely, in pi
on the module procedure-ness of dboard
: pi
already knows about it without this declaration.
Now, beyond that, because pi
is in a module there is going to be some name mangling going on for that subroutine itself. I'd solve this problem by making pi
itself a (C) interoperable procedure.
Module Fpi
IMPLICIT NONE
contains
subroutine pi(avepi, DARTS, ROUNDS) bind(C)
use, intrinsic :: iso_c_binding, only : c_double, c_int
real(c_double), intent(out) :: avepi
integer(c_int), intent(in) :: DARTS, ROUNDS
...
and then using .C
rather than .Fortran
.
You can keep pi
and dboard
in the module, and this latter needn't even be interoperable.
Try to fix the name of the fortran call inside the R function. You typed "pi" where it should be "Fpi". Also, why not bring the function to your path instead of passing a long path inside the function?
mypi <- function(DARTS, ROUNDS) {
dyn.load("./Fortran/Fpi.so")
retvals <- .Fortran("Fpi", DARTS = as.integer(DARTS) , ROUNDS = as.integer(ROUNDS), answer = as.numeric(1))
return(retvals$answer)
}
mypi(DARTS = 50000, ROUNDS = 10)