[Bug fortran/102620] New: [12 Regression] ICE in gfc_get_array_span, at fortran/trans-array.c:865

[Bug fortran/102620] New: [12 Regression] ICE in gfc_get_array_span, at fortran/trans-array.c:865

[gscfq@t-online.de]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102620

            Bug ID: 102620
           Summary: [12 Regression] ICE in gfc_get_array_span, at
                    fortran/trans-array.c:865
           Product: gcc
           Version: 12.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: fortran
          Assignee: unassigned at gcc dot gnu.org
          Reporter: gscfq@t-online.de
  Target Milestone: ---

The result of len(c) or c%len is always a scalar.
It can't be an argument of size(x) where x must be an array.
Issue started between 20210530 and 20210606 :


$ cat z1.f90
program p
   character(3), allocatable :: y(:)
   y = ['abc', 'xyz']
   call s(y)
contains
   subroutine s(x)
      character(*) :: x(:)
      integer :: i
      print *, [(len_trim(x(i)), i=1,size(x%len))]
   end
end


$ cat z1b.f90
program p
   character(:), allocatable :: y(:)
   y = ['abc', 'xyz']
   call s(y)
contains
   subroutine s(x)
      character(*) :: x(:)
      integer :: i
      print *, [(len_trim(x(i)), i=1,size(x%len))]
   end
end


$ gfortran-12-20210530 -c z1.f90
$
$ gfortran-12-20211003 -c z1.f90
z1.f90:9:50:

    9 |       print *, [(len_trim(x(i)), i=1,size(x%len))]
      |                                                  1
internal compiler error: in gfc_get_array_span, at fortran/trans-array.c:865
0x84d8aa gfc_get_array_span(tree_node*, gfc_expr*)
        ../../gcc/fortran/trans-array.c:865
0x8597c0 gfc_conv_expr_descriptor(gfc_se*, gfc_expr*)
        ../../gcc/fortran/trans-array.c:7739
0x88ce8f gfc_conv_intrinsic_size
        ../../gcc/fortran/trans-intrinsic.c:8064
0x8a488a gfc_conv_intrinsic_function(gfc_se*, gfc_expr*)
        ../../gcc/fortran/trans-intrinsic.c:10718
0x87956a gfc_conv_expr(gfc_se*, gfc_expr*)
        ../../gcc/fortran/trans-expr.c:9066
0x87c443 gfc_conv_expr_val(gfc_se*, gfc_expr*)
        ../../gcc/fortran/trans-expr.c:9119
0x857580 gfc_trans_array_constructor_value
        ../../gcc/fortran/trans-array.c:1942
0x85892a trans_array_constructor
        ../../gcc/fortran/trans-array.c:2769
0x85892a gfc_add_loop_ss_code
        ../../gcc/fortran/trans-array.c:3049
0x858ff5 gfc_conv_loop_setup(gfc_loopinfo*, locus*)
        ../../gcc/fortran/trans-array.c:5293
0x8aa9e5 gfc_trans_transfer(gfc_code*)
        ../../gcc/fortran/trans-io.c:2646
0x84aee7 trans_code
        ../../gcc/fortran/trans.c:2138
0x8a836e build_dt
        ../../gcc/fortran/trans-io.c:2026
0x84aec7 trans_code
        ../../gcc/fortran/trans.c:2110
0x871974 gfc_generate_function_code(gfc_namespace*)
        ../../gcc/fortran/trans-decl.c:6905
0x871774 gfc_generate_contained_functions
        ../../gcc/fortran/trans-decl.c:5899
0x871774 gfc_generate_function_code(gfc_namespace*)
        ../../gcc/fortran/trans-decl.c:6837
0x7f77b6 translate_all_program_units
        ../../gcc/fortran/parse.c:6572
0x7f77b6 gfc_parse_file()
        ../../gcc/fortran/parse.c:6841
0x843f3f gfc_be_parse_file
        ../../gcc/fortran/f95-lang.c:216

[Bug fortran/102620] [12 Regression] ICE in gfc_get_array_span, at fortran/trans-array.c:865

[gscfq@t-online.de]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102620

G. Steinmetz <gscfq@t-online.de> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |ice-on-invalid-code

--- Comment #1 from G. Steinmetz <gscfq@t-online.de> ---

This variant compiles and runs, but should also give an error :


$ cat z2.f90
program p
   character(3), allocatable :: y(:)
   y = ['abc', 'xyz']
   call s(y)
contains
   subroutine s(x)
      character(*), allocatable :: x(:)
      integer :: i
      print *, [(len_trim(x(i)), i=1,size(x%len))]
   end
end


$ gfortran-12-20211003 z2.f90 && ./a.out
           3           3
$

[Bug fortran/102620] [12 Regression] ICE in gfc_get_array_span, at fortran/trans-array.c:865

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102620

anlauf at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
                 CC|                            |anlauf at gcc dot gnu.org
   Last reconfirmed|                            |2021-10-05
     Ever confirmed|0                           |1

--- Comment #2 from anlauf at gcc dot gnu.org ---
Confirmed.

Reduced testcase:

subroutine s(x)
  character(*) :: x(:)
  print *, size (x%len)
end

[Bug fortran/102620] [12 Regression] ICE in gfc_get_array_span, at fortran/trans-array.c:865

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102620

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P4
   Target Milestone|---                         |12.0

[Bug fortran/102620] [12 Regression] ICE in gfc_get_array_span, at fortran/trans-array.c:865 since r12-1233-gd514626ee2566c68

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102620

Martin Liška <marxin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[12 Regression] ICE in      |[12 Regression] ICE in
                   |gfc_get_array_span, at      |gfc_get_array_span, at
                   |fortran/trans-array.c:865   |fortran/trans-array.c:865
                   |                            |since
                   |                            |r12-1233-gd514626ee2566c68
                 CC|                            |jrfsousa at gcc dot gnu.org,
                   |                            |marxin at gcc dot gnu.org

--- Comment #3 from Martin Liška <marxin at gcc dot gnu.org> ---
Confirmed, started with r12-1233-gd514626ee2566c68.

[Bug fortran/102620] [12/13 Regression] ICE in gfc_get_array_span, at fortran/trans-array.c:865 since r12-1233-gd514626ee2566c68

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102620

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|12.0                        |12.2

--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
GCC 12.1 is being released, retargeting bugs to GCC 12.2.

[Bug fortran/102620] [12/13/14 Regression] ICE in gfc_get_array_span, at fortran/trans-array.c:865 since r12-1233-gd514626ee2566c68

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102620

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|12.3                        |12.4

--- Comment #6 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 12.3 is being released, retargeting bugs to GCC 12.4.

[Bug fortran/102620] [12/13/14 Regression] ICE in gfc_get_array_span, at fortran/trans-array.c:865 since r12-1233-gd514626ee2566c68

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102620

anlauf at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |WAITING

--- Comment #7 from anlauf at gcc dot gnu.org ---
Apparently fixed on mainline for gcc-14 by r14-2461-gf9182da3213aa5.

[Bug fortran/102620] [12 Regression] ICE in gfc_get_array_span, at fortran/trans-array.c:865 since r12-1233-gd514626ee2566c68

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102620

anlauf at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[12/13 Regression] ICE in   |[12 Regression] ICE in
                   |gfc_get_array_span, at      |gfc_get_array_span, at
                   |fortran/trans-array.c:865   |fortran/trans-array.c:865
                   |since                       |since
                   |r12-1233-gd514626ee2566c68  |r12-1233-gd514626ee2566c68

--- Comment #8 from anlauf at gcc dot gnu.org ---
I get the same behavior at r13-8559 as 14-mainline.  There seems to be
another commit that fixed it independently.

Removing 13-branch from the regression list.

[Bug fortran/102620] [12 Regression] ICE in gfc_get_array_span, at fortran/trans-array.c:865 since r12-1233-gd514626ee2566c68

[pault at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102620

Paul Thomas <pault at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |pault at gcc dot gnu.org

--- Comment #9 from Paul Thomas <pault at gcc dot gnu.org> ---
(In reply to anlauf from comment #8)
> I get the same behavior at r13-8559 as 14-mainline.  There seems to be
> another commit that fixed it independently.
> 
> Removing 13-branch from the regression list.

Mark as fixed or backport fixes?

Cheers

Paul

[Bug fortran/102620] [12 Regression] ICE in gfc_get_array_span, at fortran/trans-array.c:865 since r12-1233-gd514626ee2566c68

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102620

--- Comment #10 from anlauf at gcc dot gnu.org ---
(In reply to Paul Thomas from comment #9)
> (In reply to anlauf from comment #8)
> > I get the same behavior at r13-8559 as 14-mainline.  There seems to be
> > another commit that fixed it independently.
> > 
> > Removing 13-branch from the regression list.
> 
> Mark as fixed or backport fixes?

Either I did something wrong, or the bug reappeared on 13-branch...

Anyway, I tried backporting Andre's patch to 13- and 12-branch.
Works fine and regtests fine.

How to proceed?

I can push those changes, so that we are finally done with this PR.

[Bug fortran/102620] [12 Regression] ICE in gfc_get_array_span, at fortran/trans-array.c:865 since r12-1233-gd514626ee2566c68

[pault at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102620

--- Comment #11 from Paul Thomas <pault at gcc dot gnu.org> ---
(In reply to anlauf from comment #10)
> (In reply to Paul Thomas from comment #9)
> > (In reply to anlauf from comment #8)
> > > I get the same behavior at r13-8559 as 14-mainline.  There seems to be
> > > another commit that fixed it independently.
> > > 
> > > Removing 13-branch from the regression list.
> > 
> > Mark as fixed or backport fixes?
> 
> Either I did something wrong, or the bug reappeared on 13-branch...
> 
> Anyway, I tried backporting Andre's patch to 13- and 12-branch.
> Works fine and regtests fine.
> 
> How to proceed?
> 
> I can push those changes, so that we are finally done with this PR.

Hi Harald,

It would be splendid if you would backport the patch. In the last week or so, I
have built up quite a list of backports to do, which I will attend to over the
weekend.

We are down from 105 regressions on 26th March to 94 now, of which 13 are now
fixed on mainline. Since there are still some P1 regressions, I have been
prowling around looking for more low hanging edibles while there is still time.

Regards

Paul

[Bug fortran/102620] [12 Regression] ICE in gfc_get_array_span, at fortran/trans-array.c:865 since r12-1233-gd514626ee2566c68

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102620

--- Comment #12 from anlauf at gcc dot gnu.org ---
(In reply to Paul Thomas from comment #11)
> It would be splendid if you would backport the patch. In the last week or
> so, I have built up quite a list of backports to do, which I will attend to
> over the weekend.

Hi Paul,

I've checked and found that your r14-1082-g842a432b022383 also needs to
be backported to get all testcases here fixed on 12-branch.

When regtesting completes successfully, I will push it.

[Bug fortran/102620] [12 Regression] ICE in gfc_get_array_span, at fortran/trans-array.c:865 since r12-1233-gd514626ee2566c68

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102620

anlauf at gcc dot gnu.org changed:

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

--- Comment #13 from anlauf at gcc dot gnu.org ---
Also fixed on 12-branch:

commit r12-10398-gb482968801158116dd8ba3b15a4c29143b2a423a
Author: Paul Thomas <pault@gcc.gnu.org>
Date:   Tue May 23 06:46:37 2023 +0100

    Fortran: Fix assumed length chars and len inquiry [PR103716]

    2023-05-23  Paul Thomas  <pault@gcc.gnu.org>

    gcc/fortran
            PR fortran/103716
            * resolve.cc (gfc_resolve_ref): Conversion of array_ref into an
            element should be done for all characters without a len expr,
            not just deferred lens, and for integer expressions.
            * trans-expr.cc (conv_inquiry): For len and kind inquiry refs,
            set the se string_length to NULL_TREE.

    gcc/testsuite/
            PR fortran/103716
            * gfortran.dg/pr103716.f90 : New test.

    (cherry picked from commit 842a432b02238361ecc601d301ac400a7f30f4fa)


The ICE is gone.  A remaining accepts-invalid on 11-branch could be fixed
by backporting the above if needed.

Thanks for the report to Gerhard, and to Paul for the above patch.  Closing.