This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
[Patch, libgFortran] Increase the default precision for real(16) by one
- From: Tobias Burnus <burnus at net-b dot de>
- To: gcc-patches <gcc-patches at gcc dot gnu dot org>, "'fortran at gcc dot gnu dot org'" <fortran at gcc dot gnu dot org>
- Date: Fri, 18 Jan 2008 19:01:55 +0100
- Subject: [Patch, libgFortran] Increase the default precision for real(16) by one
Using the current settings write(*,*) x sometimes does not print all
possible digits of real(16) variables on S/380.
The result is that gfortran.dg/default_format_2.f90 FAILs. That test
case contains effectively:
x = 1.0_16
do i = 0, 200
write (s,*) x
read (s,*) y
if (y /= x) res = res + 1
x = nearest(x,-huge(x))
end do
On S/380 this fails 7 times with the current setting; for these abs(x-y)
= 9.6296497219361792652798897129246366E-0035.
The solution is to simply print one more digit by default; one
additionally has to enlarge then the size of one character string in the
test suite as it is otherwise to short. I don't know whether it solves
some of the other failures on REAL(16) systems such as Debian's SPARC;
presumably it won't help on PowerPCs where the glibc seems to have a bug
(if I read PR24685 correctly).
The following patch has been build on x86-64 by me and bootstrapped and
regtested on S/390 by Andreas Krebbel.
OK for the trunk?
Tobias
2008-01-17 Tobias Burnus <burnus@net-b.de>
* io/write.c (write_real): Increase default precision for REAL(16) by one.
2008-01-17 Tobias Burnus <burnus@net-b.de>
* gfortran.dg/large_real_kind_form_io_1.f90: Enlarge string for internal I/O.
Index: libgfortran/io/write.c
===================================================================
--- libgfortran/io/write.c (Revision 131592)
+++ libgfortran/io/write.c (Arbeitskopie)
@@ -713,8 +713,8 @@ write_real (st_parameter_dt *dtp, const
f.u.real.e = 4;
break;
case 16:
- f.u.real.w = 43;
- f.u.real.d = 34;
+ f.u.real.w = 44;
+ f.u.real.d = 35;
f.u.real.e = 4;
break;
default:
Index: gfortran.dg/large_real_kind_form_io_1.f90
===================================================================
--- gfortran.dg/large_real_kind_form_io_1.f90 (Revision 131624)
+++ gfortran.dg/large_real_kind_form_io_1.f90 (Arbeitskopie)
@@ -6,7 +6,7 @@
integer, parameter :: k = selected_real_kind (precision (0.0_8) + 1)
real(kind=k) :: a,b(2), c, eps
complex(kind=k) :: d, e, f(2), g
- character(len=180) :: tmp
+ character(len=200) :: tmp
! Test real(k) scalar and array formatted IO
eps = 10 * spacing (2.0_k) ! 10 ulp precision is enough.
b(:) = 2.0_k