| Message ID | trinity-7e0ed16a-8136-41a0-809b-f25e340f5728-1602620223221@3c-app-gmx-bap55 |
|---|---|
| State | New |
| Headers | show |
| Series | PR fortran/97408 - Diagnose non-constant KIND argument to intrinsics | expand |
On 10/13/20 10:17 PM, Harald Anlauf wrote: > The KIND argument to intrinsics must be a compile-time argument. > Improve check so that the proper diagnostics is emitted. > > > - if (!gfc_check_init_expr (k)) > + if (!gfc_check_init_expr (k) || k->expr_type == EXPR_VARIABLE) I think the real question is why is the following regarded as initialization expression: t = true; … if (gfc_check_iter_variable (e)) break; Or worded differently: If integer, parameter :: A(*) = [(i, i=1,5)] is valid, which should integer, parameter :: B(*) = [integer :: (int(i, kind=i), i=1,2)] be invalid? And, indeed, the Intel Fortran compiler does accept the code: https://godbolt.org/z/EKbTf1 (While FLANG like gfortran does not.) Thus, the first question should be whether that is valid code according to the Fortran standard or not. Tobias ----------------- Mentor Graphics (Deutschland) GmbH, Arnulfstraße 201, 80634 München / Germany Registergericht München HRB 106955, Geschäftsführer: Thomas Heurung, Alexander Walter
Hi Tobias, > > The KIND argument to intrinsics must be a compile-time argument. > > Improve check so that the proper diagnostics is emitted. > > > > > > - if (!gfc_check_init_expr (k)) > > + if (!gfc_check_init_expr (k) || k->expr_type == EXPR_VARIABLE) > > I think the real question is why is the following regarded as initialization expression: > t = true; > … > if (gfc_check_iter_variable (e)) > break; you completely lost me here. Did you accidentally delete some context? > Or worded differently: If > integer, parameter :: A(*) = [(i, i=1,5)] > is valid, which should > integer, parameter :: B(*) = [integer :: (int(i, kind=i), i=1,2)] > be invalid? Well, my copy of the F2018-FDIS says about the KIND argument to INT: "KIND (optional) shall be a scalar integer constant expression." Are you saying that (int(i, kind=i), i=1,2) is legal? It would be helpful if you explained why "i" in kind=i is a constant expression. > Thus, the first question should be whether that is valid code > according to the Fortran standard or not. Indeed. Harald
Hi Harald, On 10/14/20 8:25 PM, Harald Anlauf wrote: >> Or worded differently: If >> integer, parameter :: A(*) = [(i, i=1,5)] >> is valid, which should >> integer, parameter :: B(*) = [integer :: (int(i, kind=i), i=1,2)] >> be invalid? > Well, my copy of the F2018-FDIS says about the KIND argument to INT: > "KIND (optional) shall be a scalar integer constant expression." Which applies to "B". For "A" (PARAMETER) it states: "entity has the value specified byits constant-expr," > Are you saying that (int(i, kind=i), i=1,2) is legal? > It would be helpful if you explained why "i" in kind=i is a constant expression. I only say that it might be valid. – It would be likewise helpful if you could explain why "i" is a const-expr in "[(i, i=1,5)]" – which we agree is valid, don't we?. And what about "i" in "int(i)" for "[(kind(i), i=1,5)]"? I don't know whether it is valid – I just find it not obvious that [(i, i=1,5)] is valid and [(int(1, kind=i), i=1,1)] is not. Surely, if one first expands the array, it is valid: "[integer :: (int(i, kind=i), i=1,2)]" → "[integer :: int(1, kind=1), int(2,kind=2)]" → "[integer :: 1_1, 2_2]" → "[1,2]". In any case, gfc_check_init_expr is supposed to check for const-expr – and if that does not work, gfc_check_init_expr should be fixed or at least clearly understood when it can be used and when it cannot be used. Tobias ----------------- Mentor Graphics (Deutschland) GmbH, Arnulfstraße 201, 80634 München / Germany Registergericht München HRB 106955, Geschäftsführer: Thomas Heurung, Alexander Walter
diff --git a/gcc/fortran/check.c b/gcc/fortran/check.c index 1e64fab3401..fa795538a7c 100644 --- a/gcc/fortran/check.c +++ b/gcc/fortran/check.c @@ -646,7 +646,7 @@ kind_check (gfc_expr *k, int n, bt type) if (!scalar_check (k, n)) return false; - if (!gfc_check_init_expr (k)) + if (!gfc_check_init_expr (k) || k->expr_type == EXPR_VARIABLE) { gfc_error ("%qs argument of %qs intrinsic at %L must be a constant", gfc_current_intrinsic_arg[n]->name, gfc_current_intrinsic, diff --git a/gcc/testsuite/gfortran.dg/kind_2.f90 b/gcc/testsuite/gfortran.dg/kind_2.f90 new file mode 100644 index 00000000000..f3e5b7503ef --- /dev/null +++ b/gcc/testsuite/gfortran.dg/kind_2.f90 @@ -0,0 +1,12 @@ +! { dg-do compile } +! PR97408 - Diagnose non-constant KIND argument to intrinsics + +program p + implicit none + integer :: i + integer, parameter :: lk(1) = [ 4 ] + print *, (int (1 , lk(i)), i=1,1) ! { dg-error "must be a constant" } + print *, (real (1 , lk(i)), i=1,1) ! { dg-error "must be a constant" } + print *, (cmplx (1, kind=lk(i)), i=1,1) ! { dg-error "must be a constant" } + print *, (logical (.true., lk(i)), i=1,1) ! { dg-error "must be a constant" } +end