[Bug fortran/114019] New: allocation with source of deferred character length entity

[Bug fortran/114019] New: allocation with source of deferred character length entity

[kargl at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114019

            Bug ID: 114019
           Summary: allocation with source of deferred character length
                    entity
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: fortran
          Assignee: unassigned at gcc dot gnu.org
          Reporter: kargl at gcc dot gnu.org
  Target Milestone: ---

Found this issues with the Fujitsu testsuite.  The code 
is reduced to the minimum required.  If I had to guess,
gfc_charlen for source= is not set correctly.

!
! https://github.com/fujitsu/compiler-test-suite
! Reduced from Fortran/0069/0069_0037.f90
!
  implicit none
  character(:), pointer :: chr_pointer00
  character(5), pointer :: chr_pointer01
  character(1) :: w_character01 = "4"
  allocate(chr_pointer00, source="123"//w_character01//"56")
  allocate(chr_pointer01, source="123"//w_character01//"5")
  print *,'pass'
end

% gfcx -c 0069/0069_0037.f90
0069/0069_0037.f90:9:60:

    9 |   allocate(chr_pointer00, source="123"//w_character01//"56")
      |                                                            ^
Error: size of variable 'source.2' is too large
0069/0069_0037.f90:10:59:

   10 |   allocate(chr_pointer01, source="123"//w_character01//"5")
      |                                                           ^
Error: size of variable 'source.5' is too large
during RTL pass: expand
0069/0069_0037.f90:9:60:

    9 |   allocate(chr_pointer00, source="123"//w_character01//"56")
      |                                                            ^
internal compiler error: in lhd_incomplete_type_error, at langhooks.cc:208
0x77ff6f lhd_incomplete_type_error(unsigned int, tree_node const*, tree_node
const
        ../../gccx/gcc/langhooks.cc:208
0x13c00aa size_in_bytes_loc(unsigned int, tree_node const*)
        ../../gccx/gcc/tree.cc:3601
0x13c00aa size_in_bytes_loc(unsigned int, tree_node const*)
        ../../gccx/gcc/tree.cc:3589

[Bug fortran/114019] allocation with source of deferred character length entity

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114019

anlauf at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever confirmed|0                           |1
   Last reconfirmed|                            |2024-06-07
           Keywords|                            |ice-on-valid-code
                 CC|                            |anlauf at gcc dot gnu.org

--- Comment #1 from anlauf at gcc dot gnu.org ---
Confirmed.  Reduced testcase:

subroutine s
  character(1)          :: w = "4"
  character(:), pointer :: chr_pointer
  allocate (chr_pointer, source=        "123"//w   ) ! ICE at -O0,-Og
! allocate (chr_pointer, source=adjustl("123"//w)  ) ! ICE at -O0,-Og
! allocate (chr_pointer, source=trim   ("123"//w)  ) ! OK
! allocate (chr_pointer, source=repeat ("123"//w,1)) ! OK
end

In the indicated cases where we ICE at -O0 or -Og the length of source
is determined to be 4 at compile time, in the others it is an expression.
Setting a breakpoint in gfc_allocate_using_malloc:

(gdb) p debug_tree(size)
 <integer_cst 0x7ffff6a29030 type <integer_type 0x7ffff6a25000 sizetype>
constant 4>

I do not see anything being obviously wrong with the dump-tree...

[Bug fortran/114019] allocation with source of deferred character length entity

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114019

--- Comment #2 from anlauf at gcc dot gnu.org ---
The following - seemingly hackish - change fixes the ICE:

diff --git a/gcc/fortran/trans-stmt.cc b/gcc/fortran/trans-stmt.cc
index 93b633e212e..60275e18867 100644
--- a/gcc/fortran/trans-stmt.cc
+++ b/gcc/fortran/trans-stmt.cc
@@ -6464,7 +6464,10 @@ gfc_trans_allocate (gfc_code * code, gfc_omp_namelist
*omp_allocate)
       else if (se.expr != NULL_TREE && temp_var_needed)
        {
          tree var, desc;
-         tmp = GFC_DESCRIPTOR_TYPE_P (TREE_TYPE (se.expr)) || is_coarray ?
+         tmp = (GFC_DESCRIPTOR_TYPE_P (TREE_TYPE (se.expr))
+                || is_coarray
+                || (code->expr3->ts.type == BT_CHARACTER
+                    && code->expr3->rank == 0)) ?
                se.expr
              : build_fold_indirect_ref_loc (input_location, se.expr);

For reasons I do not see the case of source being a scalar character
expression, where we already have created se as a reference a few lines
further up in the code, is not handled by the original code.

This regtests cleanly.

Is there a better solution?

[Bug fortran/114019] allocation with source of deferred character length entity

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114019

anlauf at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Assignee|unassigned at gcc dot gnu.org      |anlauf at gcc dot gnu.org

--- Comment #3 from anlauf at gcc dot gnu.org ---
Submitted: https://gcc.gnu.org/pipermail/fortran/2024-June/060609.html

[Bug fortran/114019] allocation with source of deferred character length entity

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114019

--- Comment #4 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Harald Anlauf <anlauf@gcc.gnu.org>:

https://gcc.gnu.org/g:7682d115402743090f20aca63a3b5e6c205dedff

commit r15-1722-g7682d115402743090f20aca63a3b5e6c205dedff
Author: Harald Anlauf <anlauf@gmx.de>
Date:   Fri Jun 28 21:44:06 2024 +0200

    Fortran: fix ALLOCATE with SOURCE of deferred character length [PR114019]

    gcc/fortran/ChangeLog:

            PR fortran/114019
            * trans-stmt.cc (gfc_trans_allocate): Fix handling of case of
            scalar character expression being used for SOURCE.

    gcc/testsuite/ChangeLog:

            PR fortran/114019
            * gfortran.dg/allocate_with_source_33.f90: New test.

[Bug fortran/114019] allocation with source of deferred character length entity

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114019

--- Comment #5 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-14 branch has been updated by Harald Anlauf
<anlauf@gcc.gnu.org>:

https://gcc.gnu.org/g:603b344c07aa55f8292446e8fd28f5da9a983a21

commit r14-10364-g603b344c07aa55f8292446e8fd28f5da9a983a21
Author: Harald Anlauf <anlauf@gmx.de>
Date:   Fri Jun 28 21:44:06 2024 +0200

    Fortran: fix ALLOCATE with SOURCE of deferred character length [PR114019]

    gcc/fortran/ChangeLog:

            PR fortran/114019
            * trans-stmt.cc (gfc_trans_allocate): Fix handling of case of
            scalar character expression being used for SOURCE.

    gcc/testsuite/ChangeLog:

            PR fortran/114019
            * gfortran.dg/allocate_with_source_33.f90: New test.

    (cherry picked from commit 7682d115402743090f20aca63a3b5e6c205dedff)

[Bug fortran/114019] allocation with source of deferred character length entity

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114019

anlauf at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|NEW                         |RESOLVED

--- Comment #6 from anlauf at gcc dot gnu.org ---
Fixed on mainline for gcc-15, and backported to 14-branch.  Closing.

Might be backportable to 13-branch, but not tested.