[patch] Fortran testsuite, intrinsic_count.f90

[Jerry DeLisle]

The following patch fixes a latent error in intrinsic_count.f90 that will cause 
an end-of-file error on a read with our upcoming array I/O patches.  I confirmed 
this bug with ifort and want to get this fixed now so we don't have to mingle it 
with the upcoming patches.

OK to commit to 4.1 and 4.0?  Yes, pretty much obvious but I will ask anyway.

2005-08-29  Jerry DeLisle  <jvdelisle@verizon.net>

	* gfortran.fortran-torture/execute/intrinsic_count.f90:
	Fix array dimension to avoid correct end-of-file error.


Index: intrinsic_count.f90
===================================================================
RCS file: 
/cvs/gcc/gcc/gcc/testsuite/gfortran.fortran-torture/execute/intrinsic_count.f90,v
retrieving revision 1.3
diff -c -3 -p -r1.3 intrinsic_count.f90
*** intrinsic_count.f90 9 Apr 2005 21:38:47 -0000       1.3
--- intrinsic_count.f90 29 Aug 2005 05:17:16 -0000
***************
*** 2,8 ****
   program intrinsic_count
      implicit none
      logical(kind=4), dimension (3, 5) :: a
!    integer(kind=4), dimension (5) :: b
      integer i
      character(len=10) line

--- 2,8 ----
   program intrinsic_count
      implicit none
      logical(kind=4), dimension (3, 5) :: a
!    integer(kind=4), dimension (3) :: b
      integer i
      character(len=10) line