[Bug c++/94885] New: Functional cast from int to empty class incorrectly accepted with -std=c++2a

[Bug c++/94885] New: Functional cast from int to empty class incorrectly accepted with -std=c++2a

[ppalka at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94885

            Bug ID: 94885
           Summary: Functional cast from int to empty class incorrectly
                    accepted with -std=c++2a
           Product: gcc
           Version: 10.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: ppalka at gcc dot gnu.org
  Target Milestone: ---

The following testcase

template <typename T, typename = decltype(T(0))>
void foo();

struct base {};
struct derived : base {};

void
bar()
{
  foo<derived>();
}

is rejected by GCC 9 (with both -std=c++17 and -std=c++2a) and rejected by GCC
10 with -std=c++17, but it's accepted by GCC 10 with -std=c++2a.

[Bug c++/94885] Functional cast from int to empty class type with empty base incorrectly accepted with -std=c++2a

[mpolacek at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94885

Marek Polacek <mpolacek at gcc dot gnu.org> changed:

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

--- Comment #1 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
Accepted since my r10-6519-gac6eaa55a5199196ea0a25763114ce05333a14d3

[Bug c++/94885] Functional cast from int to empty class type with empty base incorrectly accepted with -std=c++2a

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94885

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

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

--- Comment #2 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Started with r10-6519-gac6eaa55a5199196ea0a25763114ce05333a14d3

[Bug c++/94885] Functional cast from int to empty class type with empty base incorrectly accepted with -std=c++2a

[mpolacek at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94885

Marek Polacek <mpolacek at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Assignee|unassigned at gcc dot gnu.org      |mpolacek at gcc dot gnu.org
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2020-04-30
     Ever confirmed|0                           |1

[Bug c++/94885] [10 Regression] Functional cast from int to empty class type with empty base incorrectly accepted with -std=c++2a

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94885

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |10.0
            Summary|Functional cast from int to |[10 Regression] Functional
                   |empty class type with empty |cast from int to empty
                   |base incorrectly accepted   |class type with empty base
                   |with -std=c++2a             |incorrectly accepted with
                   |                            |-std=c++2a

[Bug c++/94885] [10 Regression] Functional cast from int to empty class type with empty base incorrectly accepted with -std=c++2a

[mpolacek at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94885

--- Comment #3 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
process_init_constructor_record:
1743       if (DECL_SIZE (field) && integer_zerop (DECL_SIZE (field))
1744           && !TREE_SIDE_EFFECTS (next))
1745         /* Don't add trivial initialization of an empty base/field to the
1746            constructor, as they might not be ordered the way the back-end
1747            expects.  */
1748         continue;
but here the initializer was erroneous so skipping it made us accept invalid
code.  Patch incoming.

[Bug c++/94885] [10/11 Regression] Functional cast from int to empty class type with empty base incorrectly accepted with -std=c++2a

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94885

--- Comment #4 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Marek Polacek <mpolacek@gcc.gnu.org>:

https://gcc.gnu.org/g:30da2906ac521749aef8260fc1d942e62073f19d

commit r11-20-g30da2906ac521749aef8260fc1d942e62073f19d
Author: Marek Polacek <polacek@redhat.com>
Date:   Thu Apr 30 11:36:17 2020 -0400

    c++: Parenthesized-init of aggregates accepts invalid code [PR94885]

    Here we have (conceptually *) something like

      struct B { };
      struct D : B { };
      D(0); // invalid

    and in C++20 the ()-initialization has created a { 0 } constructor that
    it tries to initialize an object of type D with.  We should reject
    initializing an object of type B from 0, but we wrongly accept it because
    process_init_constructor_record skips initializers for empty bases/fields:
           if (DECL_SIZE (field) && integer_zerop (DECL_SIZE (field))
               && !TREE_SIDE_EFFECTS (next))
             /* Don't add trivial initialization of an empty base/field to the
                constructor, as they might not be ordered the way the back-end
                expects.  */
             continue;
    but here 'next' was error_mark_node, returned by massage_elt_init, so we
    wound up with { } which would validly value-initialize the object.

    [*] Usually digest_init in build_new_method_call_1 would detect this,
    but in this case the instance is is_dummy_object and we don't call
    digest just yet.

            PR c++/94885
            * typeck2.c (process_init_constructor_record): Return
PICFLAG_ERRONEOUS
            if an initializer element was erroneous.

            * g++.dg/cpp2a/paren-init26.C: New test.

[Bug c++/94885] [10 Regression] Functional cast from int to empty class type with empty base incorrectly accepted with -std=c++2a

[mpolacek at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94885

Marek Polacek <mpolacek at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[10/11 Regression]          |[10 Regression] Functional
                   |Functional cast from int to |cast from int to empty
                   |empty class type with empty |class type with empty base
                   |base incorrectly accepted   |incorrectly accepted with
                   |with -std=c++2a             |-std=c++2a

--- Comment #5 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
Fixed on trunk so far.

[Bug c++/94885] [10 Regression] Functional cast from int to empty class type with empty base incorrectly accepted with -std=c++2a

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94885

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|10.0                        |10.2

--- Comment #6 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
GCC 10.1 has been released.

[Bug c++/94885] [10 Regression] Functional cast from int to empty class type with empty base incorrectly accepted with -std=c++2a

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94885

--- Comment #7 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-10 branch has been updated by Marek Polacek
<mpolacek@gcc.gnu.org>:

https://gcc.gnu.org/g:ceae6a13366d9646e172fc943fe8e221b70f0920

commit r10-8124-gceae6a13366d9646e172fc943fe8e221b70f0920
Author: Marek Polacek <polacek@redhat.com>
Date:   Fri May 8 11:04:29 2020 -0400

    c++: Parenthesized-init of aggregates accepts invalid code [PR94885]

    Here we have (conceptually *) something like

      struct B { };
      struct D : B { };
      D(0); // invalid

    and in C++20 the ()-initialization has created a { 0 } constructor that
    it tries to initialize an object of type D with.  We should reject
    initializing an object of type B from 0, but we wrongly accept it because
    process_init_constructor_record skips initializers for empty bases/fields:
           if (DECL_SIZE (field) && integer_zerop (DECL_SIZE (field))
               && !TREE_SIDE_EFFECTS (next))
             /* Don't add trivial initialization of an empty base/field to the
                constructor, as they might not be ordered the way the back-end
                expects.  */
             continue;
    but here 'next' was error_mark_node, returned by massage_elt_init, so we
    wound up with { } which would validly value-initialize the object.

    [*] Usually digest_init in build_new_method_call_1 would detect this,
    but in this case the instance is is_dummy_object and we don't call
    digest just yet.

            PR c++/94885
            * typeck2.c (process_init_constructor_record): Return
PICFLAG_ERRONEOUS
            if an initializer element was erroneous.

            * g++.dg/cpp2a/paren-init26.C: New test.

[Bug c++/94885] [10 Regression] Functional cast from int to empty class type with empty base incorrectly accepted with -std=c++2a

[mpolacek at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94885

Marek Polacek <mpolacek at gcc dot gnu.org> changed:

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

--- Comment #8 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
Fixed in 10.2 too.