Re: Tests for libgomp based on OpenMP Examples 4.0.2

[Ilya Verbin <iverbin@gmail.com>]

On Wed, Jul 15, 2015 at 12:29:53 +0200, Dominique d'Humières wrote:
> > The patch replaces all FP comparisons with inequalities and epsilons
> > in those tests for libgomp.
> In libgomp/testsuite/libgomp.fortran/examples-4/simd-8.f90
> 
> integer, parameter :: EPS = 0.005
> 
> should be
> 
> real, parameter :: EPS = 0.005

Committed as obvious.


2015-07-15  Maxim Blumenthal  <maxim.blumenthal@intel.com>

	* testsuite/libgomp.fortran/examples-4/simd-8.f90: (main): Change type
	of EPS parameter from integer to real.
	* testsuite/libgomp.fortran/examples-4/task_dep-5.f90: (check): Change
	type of EPS parameter from integer to real.


diff --git a/libgomp/testsuite/libgomp.fortran/examples-4/simd-8.f90 b/libgomp/testsuite/libgomp.fortran/examples-4/simd-8.f90
index ba7b0f9..3c7869d 100644
--- a/libgomp/testsuite/libgomp.fortran/examples-4/simd-8.f90
+++ b/libgomp/testsuite/libgomp.fortran/examples-4/simd-8.f90
@@ -36,7 +36,7 @@ program simd_8f
   implicit none
   real :: pri, arr(1000), diff
   integer :: i
-  integer, parameter :: EPS = 0.005
+  real, parameter :: EPS = 0.005
 
   do i = 1, 1000
      P(i)   = i
diff --git a/libgomp/testsuite/libgomp.fortran/examples-4/task_dep-5.f90 b/libgomp/testsuite/libgomp.fortran/examples-4/task_dep-5.f90
index f12b42c..0746531 100644
--- a/libgomp/testsuite/libgomp.fortran/examples-4/task_dep-5.f90
+++ b/libgomp/testsuite/libgomp.fortran/examples-4/task_dep-5.f90
@@ -44,7 +44,7 @@ contains
   subroutine check (N, A, B)
     integer :: N
     integer :: i, j
-    integer, parameter :: EPS = 0.000001
+    real, parameter :: EPS = 0.000001
     real, dimension(N,N) :: A, B
     real :: diff
     do i = 1, N


  -- Ilya