This is the mail archive of the fortran@gcc.gnu.org mailing list for the GNU Fortran project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [patch, fortran] testsuite/gfortran.dg/intrinsics


On Sunday 25 February 2007 18:52:53 Tobias Burnus wrote:
> Daniel Franke wrote:
> > http://gcc.gnu.org/wiki/GFortranTestsuiteIntrinsics
>
> I have a question regarding the autogen generation: Is is also possible
> to check for the availability of types? I would like to test also
> integer(16), real(10), real(16), which are, however, not available on
> all systems.

subroutine = {
	name = 'exit';
	arg  = 'status';     /* optional */

	call = 'CALL exit()';
	call = 'CALL exit(status = int_1)';
	call = 'CALL exit(status = int_2)';
	call = 'CALL exit(status = int_4)';
	call = 'CALL exit(status = int_8)';
	call = 'CALL exit(status = int_large)';
};

$> make check-gfortran RUNTESTFLAGS="intrinsics.exp=exit_*" 

                === gfortran Summary ===

# of expected passes            28
# of unexpected failures        2
# of unsupported tests          2


Where the failures are the known ones (PR30933) and the unsupported testcases 
(i686-pc-linux-gnu) are these:

$> cat compile/exit_sub_5.f90
! { dg-do compile }
! { dg-require-effective-target fortran_large_int }
program generated_testcase
integer, parameter :: large_int_kind  = selected_int_kind(range(0_8) + 1)
INTEGER(kind=large_int_kind) :: int_large
CALL exit(status = int_large) ! valid
end program

$> cat link/exit_sub_6.f90
! { dg-do link }
! { dg-require-effective-target fortran_large_int }
program generated_testcase
integer, parameter :: large_int_kind  = selected_int_kind(range(0_8) + 1)
INTEGER(kind=large_int_kind) :: int_large
CALL exit(status = int_large)
end program


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]