public inbox for fortran@gcc.gnu.org
 help / color / mirror / Atom feed
* [PATCH] Fortran: handle zero-sized arrays in ctors with typespec [PR108010]
@ 2022-12-07 20:57 Harald Anlauf
  2022-12-07 21:50 ` Steve Kargl
  0 siblings, 1 reply; 2+ messages in thread
From: Harald Anlauf @ 2022-12-07 20:57 UTC (permalink / raw)
  To: fortran, gcc-patches

[-- Attachment #1: Type: text/plain, Size: 291 bytes --]

Dear all,

we need to be careful about zero-sized arrays in arithmetic
reductions (unary & binary), as we otherwise may hit a NULL
pointer dereference on valid code.

The actual fix is straightforward, see attached patch.

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

Thanks,
Harald

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: pr108010.diff --]
[-- Type: text/x-patch, Size: 4321 bytes --]

From 02a8b7308d04dc84fb13b077bd3b2fe01e15c92e Mon Sep 17 00:00:00 2001
From: Harald Anlauf <anlauf@gmx.de>
Date: Wed, 7 Dec 2022 21:50:23 +0100
Subject: [PATCH] Fortran: handle zero-sized arrays in ctors with typespec
 [PR108010]

gcc/fortran/ChangeLog:

	PR fortran/108010
	* arith.cc (reduce_unary): Handle zero-sized arrays.
	(reduce_binary_aa): Likewise.

gcc/testsuite/ChangeLog:

	PR fortran/108010
	* gfortran.dg/pr108010.f90: New test.
---
 gcc/fortran/arith.cc                   | 24 ++++++++++--
 gcc/testsuite/gfortran.dg/pr108010.f90 | 54 ++++++++++++++++++++++++++
 2 files changed, 74 insertions(+), 4 deletions(-)
 create mode 100644 gcc/testsuite/gfortran.dg/pr108010.f90

diff --git a/gcc/fortran/arith.cc b/gcc/fortran/arith.cc
index c4ab75b401c..c0d12cfad9d 100644
--- a/gcc/fortran/arith.cc
+++ b/gcc/fortran/arith.cc
@@ -1342,8 +1342,16 @@ reduce_unary (arith (*eval) (gfc_expr *, gfc_expr **), gfc_expr *op,
   else
     {
       gfc_constructor *c = gfc_constructor_first (head);
-      r = gfc_get_array_expr (c->expr->ts.type, c->expr->ts.kind,
-			      &op->where);
+      if (c == NULL)
+	{
+	  /* Handle zero-sized arrays.  */
+	  r = gfc_get_array_expr (op->ts.type, op->ts.kind, &op->where);
+	}
+      else
+	{
+	  r = gfc_get_array_expr (c->expr->ts.type, c->expr->ts.kind,
+				  &op->where);
+	}
       r->shape = gfc_copy_shape (op->shape, op->rank);
       r->rank = op->rank;
       r->value.constructor = head;
@@ -1501,8 +1509,16 @@ reduce_binary_aa (arith (*eval) (gfc_expr *, gfc_expr *, gfc_expr **),
   else
     {
       gfc_constructor *c = gfc_constructor_first (head);
-      r = gfc_get_array_expr (c->expr->ts.type, c->expr->ts.kind,
-			      &op1->where);
+      if (c == NULL)
+	{
+	  /* Handle zero-sized arrays.  */
+	  r = gfc_get_array_expr (op1->ts.type, op1->ts.kind, &op1->where);
+	}
+      else
+	{
+	  r = gfc_get_array_expr (c->expr->ts.type, c->expr->ts.kind,
+				  &op1->where);
+	}
       r->shape = gfc_copy_shape (op1->shape, op1->rank);
       r->rank = op1->rank;
       r->value.constructor = head;
diff --git a/gcc/testsuite/gfortran.dg/pr108010.f90 b/gcc/testsuite/gfortran.dg/pr108010.f90
new file mode 100644
index 00000000000..303b2b98220
--- /dev/null
+++ b/gcc/testsuite/gfortran.dg/pr108010.f90
@@ -0,0 +1,54 @@
+! { dg-do run }
+! PR fortran/108010 - ICE in reduce_unary, reduce_binary_aa
+! Contributed by G.Steinmetz
+
+program p
+  implicit none
+  print *,   + [integer :: [real ::]]
+  print *,   - [integer :: [real ::]]
+  print *, 1 + [integer :: [real ::]]
+  print *, 1 - [integer :: [real ::]]
+  print *, 2 * [integer :: [real ::]]
+  print *,   - [real :: [real ::], 2]
+  print *,   + [integer :: [real ::], 2]
+  print *,   - [integer :: [real ::], 2]
+  print *, 1 + [integer :: [real ::], 2]
+  print *, 1 - [integer :: [real ::], 2]
+  print *, 2 * [integer :: [real ::], 2]
+  print *, [integer :: [real ::]] + [integer :: [real ::]]
+  print *, [integer :: [real ::]] - [integer :: [real ::]]
+  print *, [integer :: [real ::]] * [integer :: [real ::]]
+  print *, [integer :: [real ::], 2] + [real :: [real ::], 3]
+  print *, [integer :: [real ::], 2] - [real :: [real ::], 3]
+  print *, [integer :: [real ::], 2] * [real :: [real ::], 3]
+
+  ! Validate type of resulting arrays
+  if (.not. is_int ([integer :: [real ::]]                         )) stop 1
+  if (.not. is_int ([integer :: [real ::]] + [integer :: [real ::]])) stop 2
+  if (.not. is_real([real :: [integer ::]]                         )) stop 3
+  if (.not. is_real([real :: [integer ::]] + [real :: [integer ::]])) stop 4
+  if (.not. is_real([real :: [integer ::]] + [integer :: [real ::]])) stop 5
+  if (.not. is_real([integer :: [real ::]] + [real :: [integer ::]])) stop 6
+
+contains
+
+  logical function is_int (x)
+    class(*) :: x(:)
+    select type (x)
+    type is (integer)
+       is_int = .true.
+    class default
+       is_int = .false.
+    end select
+  end function is_int
+
+  logical function is_real (x)
+    class(*) :: x(:)
+    select type (x)
+    type is (real)
+       is_real = .true.
+    class default
+       is_real = .false.
+    end select
+  end function is_real
+end
--
2.35.3


^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [PATCH] Fortran: handle zero-sized arrays in ctors with typespec [PR108010]
  2022-12-07 20:57 [PATCH] Fortran: handle zero-sized arrays in ctors with typespec [PR108010] Harald Anlauf
@ 2022-12-07 21:50 ` Steve Kargl
  0 siblings, 0 replies; 2+ messages in thread
From: Steve Kargl @ 2022-12-07 21:50 UTC (permalink / raw)
  To: Harald Anlauf via Fortran; +Cc: gcc-patches

On Wed, Dec 07, 2022 at 09:57:20PM +0100, Harald Anlauf via Fortran wrote:
> Dear all,
> 
> we need to be careful about zero-sized arrays in arithmetic
> reductions (unary & binary), as we otherwise may hit a NULL
> pointer dereference on valid code.
> 
> The actual fix is straightforward, see attached patch.
> 
> Regtested on x86_64-pc-linux-gnu.  OK for mainline?
> 

Yes.  Thanks for the patch.

-- 
Steve

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2022-12-07 21:51 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-12-07 20:57 [PATCH] Fortran: handle zero-sized arrays in ctors with typespec [PR108010] Harald Anlauf
2022-12-07 21:50 ` Steve Kargl

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).