Re: [Dev-C++] Mixed Language Programming w/Dev-Cpp

[<h.r...@gm...>]

> Hi,
> 
> Wondering if anyone has experience with mixed language programming and
> whether it can be done with Dev-Cpp?
> 
> In particular, I'm writing a mathematical simulation in C (using Dev-Cpp)
> and would like to make using of some existing numerical routines in
> Fortran.
> My C knowledge is necessity-based self-taught and so limited; don't know
> anything about Fortran.  Neither do I know much about using
> compilers/linkers/etc in command line mode (which is why I use Dev-Cpp--I
> let it worry about that stuff).
> 
> Still, I know that built-in to the gcc interface is the g77 compiler so
> there must be a way to compile the fortran source to an object with g77
> and
> then link to & call the routines from C.  Could it be as simple as added
> fortran source files to my Dev-Cpp project?  If so, any thoughts on how to
> prototype the Fortran functions & subfunctions to make them accessible to
> the C calling routines?
> 
> BTW, I've tried f2c but have been unable to make that work.  That approach
> is suboptimal anyway because the resulting C code is far less efficient
> than
> the original Fortran source.
> 
> Perhaps specifics would help.  The fortran code I'm trying to use is a
> single file ode.f  The entry point is the subroutine ode.  Here's how the
> file is laid out:
> 
> // ode.f
> 
>       subroutine ode(f,neqn,y,t,tout,relerr,abserr,iflag,work,iwork)
>       implicit real*8(a-h,o-z)
> 
> c      f -- double precision subroutine f(t,y,yp) to evaluate
> c                derivatives yp(i)=dy(i)/dt
> c      neqn -- number of equations to be integrated (integer*4)
> c      y(*) -- solution vector at t                 (real*8)
> c      t -- independent variable                    (real*8)
> c      tout -- point at which solution is desired   (real*8)
> c      relerr,abserr -- relative and absolute error tolerances for local
> c           error test (real*8).  at each step the code requires
> c             dabs(local error) .le. dabs(y)*relerr + abserr
> c           for each component of the local error and solution vectors
> c      iflag -- indicates status of integration     (integer*4)
> c      work(*)  (real*8)  -- arrays to hold information internal to
> c      iwork(*) (integer*4)    which is necessary for subsequent calls
> 
> . . . ode calls de:
>       subroutine de(f,neqn,y,t,tout,relerr,abserr,iflag,
>      1  yy,wt,p,yp,ypout,phi,alpha,beta,sig,v,w,g,phase1,psi,x,h,hold,
>      2  start,told,delsgn,ns,nornd,k,kold,isnold)
>       implicit real*8(a-h,o-z)
> 
> . . . de calls step & intrp:
>       subroutine step(x,y,f,neqn,h,eps,wt,start,
>      1  hold,k,kold,crash,phi,p,yp,psi,
>      2  alpha,beta,sig,v,w,g,phase1,ns,nornd)
>       implicit real*8(a-h,o-z)
> 
>       subroutine intrp(x,y,xout,yout,ypout,neqn,kold,phi,psi)
>       implicit real*8(a-h,o-z)
> // end ode.f
> 
> The derivative function f is written in C (and that's where I was having
> probs with the f2c route).
> 
> All help is appreciated (sorry for the long post).
> 
> Scott
> 
there are many interesting articels about this topic in the web, e.g.:
Carsten A. Arnholm
Mixed language programming using C++ and FORTRAN 77
at: http://home.online.no/~arnholm/cppf77.htm#Section1
If you use google for search you will find many others.

Horst

-- 
Horst Rößner
Postf. 1802
52205 Stolberg

ICQ ID #74808448
h.r...@gm...
h.g...@t-...

+++ GMX - Mail, Messaging & more  http://www.gmx.net +++
NEU: Mit GMX ins Internet. Rund um die Uhr für 1 ct/ Min. surfen!