gfortran bug?

[Daniel Franke]

On Wednesday 06 December 2006 21:24, Ray Nachlinger wrote:
> The following program produces strange results.

[snipped program code]

> "repeat" takes the first argument which is len integer*2
> length long and copies it num times to the last argument.
> As far a I can see, all three calls should produce 5
> .true. values. The first two calls, however, produce
> T F T T T. The last one is as I expected. Is this a bug
> or am I confused?
>
> I am using gcc version 4.1.1

$> gfortran-svn -g -Wall ray.f90 && ./a.out
 T T T T T
 T T T T T
 T T T T T
Segmentation fault

$> gfortran-svn -v
gcc version 4.3.0 20061128 (experimental)

Works here in terms of expectations - but crashes afterwards.


Slightly rewritten, making "repeat" a CONTAINed subroutine, I get:

$> gfortran-svn -g -Wall -fbounds-check ray-2.f90
ray-2.f90:7.14:

  call repeat(.true.,len_l,n,   l(1))
             1
Error: Type/rank mismatch in argument 'a' at (1)
ray-2.f90:10.14:

  call repeat(lt,len_l,n,   l(1))
             1
Error: Type/rank mismatch in argument 'a' at (1)
ray-2.f90:14.14:

  call repeat(l(1),len_l,n,   l(2))
             1
Error: Type/rank mismatch in argument 'a' at (1)


(These diagnostics are also given by 4.1.1)


$> cat ray-2.f90
PROGRAM main
  integer, parameter :: len_l = 2, n = 5
  logical, parameter :: lt = .true.
  logical l(n)

  call repeat(.true.,len_l,n,   l(1))
  print *,l

  call repeat(lt,len_l,n,   l(1))
  print *,l

  l(1) = .true.
  call repeat(l(1),len_l,n,   l(2))
  print *,l

  CONTAINS
    subroutine repeat(a,len,num,   b)
      integer num,len,k,j
      integer*2 a(num,*),b(num,*)
      do 2100 k = 1, num
        do 2100 j = 1, len
           b(j,k) = a(j,k)
      2100 continue
    end subroutine
END PROGRAM


Hope this helps?!
Regards.

	Daniel