Hi Nathan! On Sat, 25 Jul 2015 16:02:01 -0400, Nathan Sidwell <nat...@acm.org> wrote: > I've committed this to gomp4 branch. It fixes some tests that were incorrect
Hmm, I fail to see what you deem incorrect in the following two Fortran test cases? Implicit present_or_copy clauses should be added by the compiler, basically equal to your explicit present clauses. > and fail with some development I am working on. Fail in what way? I'd expect the original code still to be valid? > * testsuite/libgomp.oacc-fortran/data-2.f90: Add present clauses > to parallels. > * testsuite/libgomp.oacc-fortran/lib-14.f90: Likewise. > --- libgomp/testsuite/libgomp.oacc-fortran/data-2.f90 (revision 226189) > +++ libgomp/testsuite/libgomp.oacc-fortran/data-2.f90 (working copy) > @@ -19,7 +19,7 @@ program test > > !$acc enter data copyin (a(1:N,1:N), b(1:N,1:N)) > > - !$acc parallel > + !$acc parallel present (a(1:N,1:N), b(1:N,1:N)) > do i = 1, n > do j = 1, n > b(j,i) = a (j,i) > @@ -45,7 +45,7 @@ program test > !$acc enter data copyin (c(1:N)) create (d(1:N)) async > !$acc wait > > - !$acc parallel > + !$acc parallel present (c(1:N), d(1:N)) > do i = 1, N > d(i) = c(i) + 1 > end do > @@ -65,7 +65,7 @@ program test > !$acc enter data create (d(1:N)) wait > !$acc wait > > - !$acc parallel > + !$acc parallel present (c(1:N), d(1:N)) > do i = 1, N > d(i) = c(i) + 1 > end do > @@ -128,7 +128,7 @@ program test > if (acc_is_present (c) .eqv. .FALSE.) call abort > if (acc_is_present (d) .eqv. .FALSE.) call abort > > - !$acc parallel > + !$acc parallel present (c(0:N), d(0:N)) > do i = 1, N > c(i) = 1.0; > d(i) = 2.0; > --- libgomp/testsuite/libgomp.oacc-fortran/lib-14.f90 (revision 226189) > +++ libgomp/testsuite/libgomp.oacc-fortran/lib-14.f90 (working copy) > @@ -46,7 +46,7 @@ program main > > if (acc_is_present (h) .neqv. .TRUE.) call abort > > - !$acc parallel loop > + !$acc parallel loop present (h) > do i = 1, N > h(i) = i > end do Grüße, Thomas
signature.asc
Description: PGP signature