[PATCH] PR fortran/104619 - [10/11/12 Regression] ICE on list comprehension with default derived type constructor

[PATCH] PR fortran/104619 - [10/11/12 Regression] ICE on list comprehension with default derived type constructor

[Harald Anlauf]

[-- Attachment #1: Type: text/plain, Size: 361 bytes --]

Dear Fortranners,

a recently introduced shape validation for an array constructor
against the declared shape of a DT component failed to punt if
the shape of the constructor cannot be determined at compile time.

Suggested solution: skip the shape check in those cases.

Regtested on x86_64-pc-linux-gnu.  OK for mainline / affected branches?

Thanks,
Harald


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-Fortran-skip-compile-time-shape-check-if-constructor.patch --]
[-- Type: text/x-patch, Size: 2270 bytes --]

From 356d1ab6ddb58559d3641f169f44fd24f7cb00d1 Mon Sep 17 00:00:00 2001
From: Harald Anlauf <anlauf@gmx.de>
Date: Mon, 21 Feb 2022 22:49:05 +0100
Subject: [PATCH] Fortran: skip compile-time shape check if constructor shape
 is not known

gcc/fortran/ChangeLog:

	PR fortran/104619
	* resolve.cc (resolve_structure_cons): Skip shape check if shape
	of constructor cannot be determined at compile time.

gcc/testsuite/ChangeLog:

	PR fortran/104619
	* gfortran.dg/derived_constructor_comps_7.f90: New test.
---
 gcc/fortran/resolve.cc                        |  2 ++
 .../derived_constructor_comps_7.f90           | 28 +++++++++++++++++++
 2 files changed, 30 insertions(+)
 create mode 100644 gcc/testsuite/gfortran.dg/derived_constructor_comps_7.f90

diff --git a/gcc/fortran/resolve.cc b/gcc/fortran/resolve.cc
index 266e41e25b1..451bc97df43 100644
--- a/gcc/fortran/resolve.cc
+++ b/gcc/fortran/resolve.cc
@@ -1472,6 +1472,8 @@ resolve_structure_cons (gfc_expr *expr, int init)
 		  t = false;
 		  break;
 		};
+	      if (cons->expr->shape == NULL)
+		continue;
 	      mpz_set_ui (len, 1);
 	      mpz_add (len, len, comp->as->upper[n]->value.integer);
 	      mpz_sub (len, len, comp->as->lower[n]->value.integer);
diff --git a/gcc/testsuite/gfortran.dg/derived_constructor_comps_7.f90 b/gcc/testsuite/gfortran.dg/derived_constructor_comps_7.f90
new file mode 100644
index 00000000000..238ac3d9f26
--- /dev/null
+++ b/gcc/testsuite/gfortran.dg/derived_constructor_comps_7.f90
@@ -0,0 +1,28 @@
+! { dg-do run }
+! PR fortran/104619
+
+module m
+  implicit none
+  type :: item
+     real :: x
+  end type item
+  type :: container
+     type(item) :: items(3)
+  end type container
+end module
+
+program p
+  use m
+  implicit none
+  type(item), allocatable :: items(:)
+  type(container) :: c
+  integer         :: i, n
+  items = [item(3.0), item(4.0), item(5.0)]
+  c = container(items=[(items(i), i = 1, size(items))])
+  if (any (c%items% x /= items% x)) stop 1
+  n = size (items)
+  c = container(items=[(items(i), i = 1, n)])
+  if (any (c%items% x /= items% x)) stop 2
+  c = container(items=[(items(i), i = 1, 3)])
+  if (any (c%items% x /= items% x)) stop 3
+end program
--
2.34.1

Re: [PATCH] PR fortran/104619 - [10/11/12 Regression] ICE on list comprehension with default derived type constructor

[Thomas Koenig]

Hi Harald,

> a recently introduced shape validation for an array constructor
> against the declared shape of a DT component failed to punt if
> the shape of the constructor cannot be determined at compile time.
> 
> Suggested solution: skip the shape check in those cases.
> 
> Regtested on x86_64-pc-linux-gnu.  OK for mainline / affected branches?

Looks good to me.

Thanks for the patch!

Best regards

	Thomas