Different warning with *.f and *.f90 suffix (was Compilation and run of the FDS code)

[Steve Kargl]

On Tue, Apr 18, 2006 at 10:13:09PM +0200, Dominique Dhumieres wrote:
> I have now at least a partial understanding of what happened with
> the warnings.  I have reduced it to:
> 
> [karma] f90/bug% cat read_bndf_1.f90
>       SUBROUTINE READ_BNDF
> !
>       CHARACTER(30) QUANTITY
>       NAMELIST /BNDF/ DTSAM,QUANTITY,FYI
> !
>       BNDFLOOP: DO N=1,NBF
>       READ(LU5,BNDF,END=243,ERR=244)
> 
>       SEARCH: DO ND=-NDATA,NDATA
>       ENDDO SEARCH
> 
>   244 ENDDO BNDFLOOP
>   243 REWIND(LU5)
> 
>       END SUBROUTINE READ_BNDF
> 
> Then I get:
> 
> [karma] f90/bug% gfortran -c read_bndf_1.f90
>  In file read_bndf_1.f90:7
> 
>       READ(LU5,BNDF,END=243,ERR=244)
>                                    1
>  In file read_bndf_1.f90:12
> 
>   244 ENDDO BNDFLOOP
>                    2
> Warning: Obsolete: GOTO at (1) jumps to END of construct at (2)
> [karma] f90/bug% cp read_bndf_1.f90 read_bndf_1.f
> [karma] f90/bug% gfortran -c read_bndf_1.f
>  In file read_bndf_1.f:7
> 
>       READ(LU5,BNDF,END=243,ERR=244)                                    
>                                                                        1
>  In file read_bndf_1.f:12
> 
>                     
>                    2
> Warning: Obsolete: GOTO at (1) jumps to END of construct at (2)
> 
> So the warning is at it should be with the f90 suffix, but not with the
> f one. Note that the 'SEARCH' do loop is required to get the warning
> for reasons I don't understand.

I don't understand your comment about the 'SEARCH' loop.

-- 
Steve