[Bug fortran/114739] New: [14 Regression] ice in gfc_find_derived_types, at fortran/symbol.cc:2458

[Bug fortran/114739] New: [14 Regression] ice in gfc_find_derived_types, at fortran/symbol.cc:2458

[dcb314 at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114739

            Bug ID: 114739
           Summary: [14 Regression] ice in gfc_find_derived_types, at
                    fortran/symbol.cc:2458
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: fortran
          Assignee: unassigned at gcc dot gnu.org
          Reporter: dcb314 at hotmail dot com
  Target Milestone: ---

From the flang testsuite at

https://github.com/llvm/llvm-project/tree/main/flang/test

file ./Semantics/symbol07.f90, does this with gfortran 13.2:

$ ~/gcc/results.13.2.asan.ubsan/bin/gfortran -c -w ./Semantics/symbol07.f90
./Semantics/symbol07.f90:28:5:

   28 |  z2%re = x
      |     1
Error: Symbol ‘z2’ at (1) has no IMPLICIT type
./Semantics/symbol07.f90:31:5:

   31 |  z2%im = y
      |     1
Error: Symbol ‘z2’ at (1) has no IMPLICIT type
$

and this with yesterday's gfortran:

$ ~/gcc/results.20240415.asan.ubsan/bin/gfortran -c -w ./Semantics/symbol07.f90
f951: internal compiler error: in gfc_find_derived_types, at
fortran/symbol.cc:2458
0x6f3ca3 gfc_find_derived_types(gfc_symbol*, gfc_namespace*, char const*, bool)
        ../../trunk.20210101/gcc/fortran/symbol.cc:2458
0x9e1686 gfc_match_varspec(gfc_expr*, int, bool, bool)
        ../../trunk.20210101/gcc/fortran/primary.cc:2246
0x9e1916 match_variable
        ../../trunk.20210101/gcc/fortran/primary.cc:4309
0x99a364 gfc_match(char const*, ...)
        ../../trunk.20210101/gcc/fortran/match.cc:1144

[Bug fortran/114739] [14 Regression] ice in gfc_find_derived_types, at fortran/symbol.cc:2458

[kargls at comcast dot net]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114739

kargls at comcast dot net changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |kargls at comcast dot net

--- Comment #1 from kargls at comcast dot net ---
Please attach a the source code.

[Bug fortran/114739] [14 Regression] ice in gfc_find_derived_types, at fortran/symbol.cc:2458

[dcb314 at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114739

--- Comment #2 from David Binderman <dcb314 at hotmail dot com> ---
Created attachment 57959
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=57959&action=edit
F90 source code

[Bug fortran/114739] [14 Regression] ice in gfc_find_derived_types, at fortran/symbol.cc:2458

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114739

anlauf at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |ice-on-valid-code,
                   |                            |rejects-valid
             Status|UNCONFIRMED                 |NEW
     Ever confirmed|0                           |1
           Priority|P3                          |P4
   Last reconfirmed|                            |2024-04-16

--- Comment #3 from anlauf at gcc dot gnu.org ---
We fail to figure out that z2 is complex.

Adding
  complex z2
avoid the issues.

Not really a regression except for the ICE, since we falsely rejected the code
up to 13.

[Bug fortran/114739] [14 Regression] ice in gfc_find_derived_types, at fortran/symbol.cc:2458

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114739

--- Comment #4 from anlauf at gcc dot gnu.org ---
Reduced testcase:

program main
  implicit complex(z)
  z2%re = 1.
  z2%im = 2.
end

[Bug fortran/114739] [14 Regression] ice in gfc_find_derived_types, at fortran/symbol.cc:2458

[anlauf at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114739

anlauf at gcc dot gnu.org changed:

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

--- Comment #5 from anlauf at gcc dot gnu.org ---
We're running into code in primary.cc modified at r14-9489-g3fd46d859cda10
which checks for inquiry references.

CC'ing Paul.

[Bug fortran/114739] [14 Regression] ice in gfc_find_derived_types, at fortran/symbol.cc:2458

[pault at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114739

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

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

--- Comment #6 from Paul Thomas <pault at gcc dot gnu.org> ---
Hi David and Harald,

Thanks for the heads up.

I am within minutes of posting a fix on the list.

Paul

[Bug fortran/114739] [14 Regression] ice in gfc_find_derived_types, at fortran/symbol.cc:2458

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114739

--- Comment #7 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Paul Thomas <pault@gcc.gnu.org>:

https://gcc.gnu.org/g:e243d0feafa533141ef7e23820d5cc60cf33204a

commit r14-10030-ge243d0feafa533141ef7e23820d5cc60cf33204a
Author: Paul Thomas <pault@gcc.gnu.org>
Date:   Thu Apr 18 18:07:25 2024 +0100

    Fortran: Fix ICE and clear incorrect error messages [PR114739]

    2024-04-18  Paul Thomas  <pault@gcc.gnu.org>

    gcc/fortran
            PR fortran/114739
            * primary.cc (gfc_match_varspec): Check for default type before
            checking for derived types with the right component name.

    gcc/testsuite/
            PR fortran/114739
            * gfortran.dg/pr114739.f90: New test.
            * gfortran.dg/derived_comp_array_ref_8.f90: Add 'implicit none'
            for consistency with expected error message.
            * gfortran.dg/nullify_4.f90: ditto
            * gfortran.dg/pointer_init_6.f90: ditto
            * gfortran.dg/pr107397.f90: ditto
            * gfortran.dg/pr88138.f90: ditto

[Bug fortran/114739] [14 Regression] ice in gfc_find_derived_types, at fortran/symbol.cc:2458

[pault at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114739

--- Comment #8 from Paul Thomas <pault at gcc dot gnu.org> ---
Thanks for the report. Fixed on mainline.

The incorrect errors will have to be fixed on 13-branch at very least.
Unfortunately, this is not a question of using a backport but I will get onto
it right away.

Regards

Paul

[Bug fortran/114739] [14 Regression] ice in gfc_find_derived_types, at fortran/symbol.cc:2458

[pault at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114739

--- Comment #9 from Paul Thomas <pault at gcc dot gnu.org> ---

> The incorrect errors will have to be fixed on 13-branch at very least.
> Unfortunately, this is not a question of using a backport but I will get
> onto it right away.
> 


I have applied the backport manually and it works fine in eliminating the
incorrect error. I'll apply to 12- and 13-branches in a couple of weeks.

Cheers

Paul

[Bug fortran/114739] Ensure no IMPLICIT type errors appear when they should for inquiry references

[pault at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114739

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[14 Regression] ice in      |Ensure no IMPLICIT type
                   |gfc_find_derived_types, at  |errors appear when they
                   |fortran/symbol.cc:2458      |should for inquiry
                   |                            |references

--- Comment #10 from Paul Thomas <pault at gcc dot gnu.org> ---
Changed the summary line

[Bug fortran/114739] Ensure no IMPLICIT type errors appear when they should for inquiry references

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114739

--- Comment #11 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-13 branch has been updated by Paul Thomas <pault@gcc.gnu.org>:

https://gcc.gnu.org/g:094f8a36dffea52fe6a64596203b82648a3f6121

commit r13-8689-g094f8a36dffea52fe6a64596203b82648a3f6121
Author: Paul Thomas <pault@gcc.gnu.org>
Date:   Mon May 6 08:21:14 2024 +0100

    Fortran: Fix ICE and clear incorrect error messages [PR114739]

    2024-05-06  Paul Thomas  <pault@gcc.gnu.org>

    gcc/fortran
            PR fortran/114739
            * primary.cc (gfc_match_varspec): Check for default type before
            checking for derived types with the right component name.

    gcc/testsuite/
            PR fortran/114739
            * gfortran.dg/pr114739.f90: New test.
            * gfortran.dg/derived_comp_array_ref_8.f90: Add 'implicit none'
            for consistency with expected error message.
            * gfortran.dg/nullify_4.f90: ditto
            * gfortran.dg/pointer_init_6.f90: ditto
            * gfortran.dg/pr107397.f90: ditto
            * gfortran.dg/pr88138.f90: ditto

[Bug fortran/114739] Ensure no IMPLICIT type errors appear when they should for inquiry references

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114739

--- Comment #12 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-12 branch has been updated by Paul Thomas <pault@gcc.gnu.org>:

https://gcc.gnu.org/g:3a5acd2583056e8cd0e5fda83e7c34be65415c62

commit r12-10415-g3a5acd2583056e8cd0e5fda83e7c34be65415c62
Author: Paul Thomas <pault@gcc.gnu.org>
Date:   Mon May 6 08:21:14 2024 +0100

    Fortran: Fix ICE and clear incorrect error messages [PR114739]

    2024-05-06  Paul Thomas  <pault@gcc.gnu.org>

    gcc/fortran
            PR fortran/114739
            * primary.cc (gfc_match_varspec): Check for default type before
            checking for derived types with the right component name.

    gcc/testsuite/
            PR fortran/114739
            * gfortran.dg/pr114739.f90: New test.
            * gfortran.dg/derived_comp_array_ref_8.f90: Add 'implicit none'
            for consistency with expected error message.
            * gfortran.dg/nullify_4.f90: ditto
            * gfortran.dg/pointer_init_6.f90: ditto
            * gfortran.dg/pr107397.f90: ditto
            * gfortran.dg/pr88138.f90: ditto

    (cherry picked from commit 094f8a36dffea52fe6a64596203b82648a3f6121)

[Bug fortran/114739] Ensure no IMPLICIT type errors appear when they should for inquiry references

[pault at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114739

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

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

--- Comment #13 from Paul Thomas <pault at gcc dot gnu.org> ---
Backported to 12- and 13-branches.

Closing

Paul