[Bug c++/112439] New: Modification of a member overlapping with a [[no_unique_address]] member in the constructor is incorrectly rejected

[Bug c++/112439] New: Modification of a member overlapping with a [[no_unique_address]] member in the constructor is incorrectly rejected

[de34 at live dot cn]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112439

            Bug ID: 112439
           Summary: Modification of a member overlapping with a
                    [[no_unique_address]] member in the constructor is
                    incorrectly rejected
           Product: gcc
           Version: 13.2.1
            Status: UNCONFIRMED
          Keywords: rejects-valid
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: de34 at live dot cn
  Target Milestone: ---

GCC starts (incorrectly) rejecting the following code snipped since GCC13
(https://godbolt.org/z/faqMahz34). It's wrong to emit an error for this because
c is not const in the constructor body.

```
struct Empty {};

class Foo {
public:
    constexpr Foo(int x, Empty y, int z) : a(x), b(y)
    {
        c = z;
    }

private:
    int a{};
    [[no_unique_address]] Empty b{};
    [[no_unique_address]] int c{};
};

constexpr Foo r{1, {}, 3};
```

It seems that the code is correctly accepted if b and c are made not to
overlap.

[Bug c++/112439] Modification of a member overlapping with a [[no_unique_address]] member in the constructor is incorrectly rejected in constant evaluation

[redi at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112439

Jonathan Wakely <redi at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2023-11-08

[Bug c++/112439] [13/14 Regression] Modification of a member overlapping with a [[no_unique_address]] member in the constructor is incorrectly rejected in constant evaluation

[ppalka at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112439

Patrick Palka <ppalka at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jason at gcc dot gnu.org,
                   |                            |ppalka at gcc dot gnu.org
      Known to work|                            |12.3.0
      Known to fail|                            |13.2.0, 14.0
   Target Milestone|---                         |13.3
            Summary|Modification of a member    |[13/14 Regression]
                   |overlapping with a          |Modification of a member
                   |[[no_unique_address]]       |overlapping with a
                   |member in the constructor   |[[no_unique_address]]
                   |is incorrectly rejected in  |member in the constructor
                   |constant evaluation         |is incorrectly rejected in
                   |                            |constant evaluation

--- Comment #1 from Patrick Palka <ppalka at gcc dot gnu.org> ---
Started with r13-160-g967cdbe6629653

[Bug c++/112439] [13/14 Regression] Modification of a member overlapping with a [[no_unique_address]] member in the constructor is incorrectly rejected in constant evaluation since r13-160-g967cdbe6629653

[jason at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112439

Jason Merrill <jason at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
           Assignee|unassigned at gcc dot gnu.org      |jason at gcc dot gnu.org

[Bug c++/112439] [13/14 Regression] Modification of a member overlapping with a [[no_unique_address]] member in the constructor is incorrectly rejected in constant evaluation since r13-160-g967cdbe6629653

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112439

--- Comment #2 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The trunk branch has been updated by Jason Merrill <jason@gcc.gnu.org>:

https://gcc.gnu.org/g:f4998609908e4926fc095ce97cb84b187294fd1d

commit r14-8727-gf4998609908e4926fc095ce97cb84b187294fd1d
Author: Jason Merrill <jason@redhat.com>
Date:   Thu Feb 1 17:23:53 2024 -0500

    c++: no_unique_address and constexpr [PR112439]

    Here, because we don't build a CONSTRUCTOR for an empty base, we were
    wrongly marking the Foo CONSTRUCTOR as complete after initializing the
Empty
    member.  Fixed by checking empty_base here as well.

            PR c++/112439

    gcc/cp/ChangeLog:

            * constexpr.cc (cxx_eval_store_expression): Check empty_base
            before marking a CONSTRUCTOR readonly.

    gcc/testsuite/ChangeLog:

            * g++.dg/cpp2a/no_unique_address15.C: New test.

[Bug c++/112439] [13/14 Regression] Modification of a member overlapping with a [[no_unique_address]] member in the constructor is incorrectly rejected in constant evaluation since r13-160-g967cdbe6629653

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112439

--- Comment #3 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-13 branch has been updated by Jason Merrill
<jason@gcc.gnu.org>:

https://gcc.gnu.org/g:8bae2c531d2ea8f7872ed41f273b6fabaa6a9965

commit r13-8271-g8bae2c531d2ea8f7872ed41f273b6fabaa6a9965
Author: Jason Merrill <jason@redhat.com>
Date:   Thu Feb 1 17:23:53 2024 -0500

    c++: no_unique_address and constexpr [PR112439]

    Here, because we don't build a CONSTRUCTOR for an empty base, we were
    wrongly marking the Foo CONSTRUCTOR as complete after initializing the
Empty
    member.  Fixed by checking empty_base here as well.

            PR c++/112439

    gcc/cp/ChangeLog:

            * constexpr.cc (cxx_eval_store_expression): Check empty_base
            before marking a CONSTRUCTOR readonly.

    gcc/testsuite/ChangeLog:

            * g++.dg/cpp2a/no_unique_address15.C: New test.

    (cherry picked from commit f4998609908e4926fc095ce97cb84b187294fd1d)

[Bug c++/112439] [13/14 Regression] Modification of a member overlapping with a [[no_unique_address]] member in the constructor is incorrectly rejected in constant evaluation since r13-160-g967cdbe6629653

[jason at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112439

Jason Merrill <jason at gcc dot gnu.org> changed:

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

--- Comment #4 from Jason Merrill <jason at gcc dot gnu.org> ---
Fixed for 13.3/14.