public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug libstdc++/111948] New: subrange modifies a const size object
@ 2023-10-24  5:51 hewillk at gmail dot com
  2023-10-24  5:55 ` [Bug libstdc++/111948] " hewillk at gmail dot com
                   ` (11 more replies)
  0 siblings, 12 replies; 13+ messages in thread
From: hewillk at gmail dot com @ 2023-10-24  5:51 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111948

            Bug ID: 111948
           Summary: subrange modifies a const size object
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: libstdc++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: hewillk at gmail dot com
  Target Milestone: ---

https://godbolt.org/z/Gf4zWP3qT

testcase:

#include <ranges>

constexpr auto r = std::ranges::subrange(std::views::iota(0), 5);
static_assert(std::ranges::distance(r));

^ permalink raw reply	[flat|nested] 13+ messages in thread

* [Bug libstdc++/111948] subrange modifies a const size object
  2023-10-24  5:51 [Bug libstdc++/111948] New: subrange modifies a const size object hewillk at gmail dot com
@ 2023-10-24  5:55 ` hewillk at gmail dot com
  2023-10-24 16:15 ` redi at gcc dot gnu.org
                   ` (10 subsequent siblings)
  11 siblings, 0 replies; 13+ messages in thread
From: hewillk at gmail dot com @ 2023-10-24  5:55 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111948

--- Comment #1 from 康桓瑋 <hewillk at gmail dot com> ---
This is the cause:

      constexpr
      subrange(__detail::__convertible_to_non_slicing<_It> auto __i, _Sent __s,
               __size_type __n)
      noexcept(is_nothrow_constructible_v<_It, decltype(__i)>
               && is_nothrow_constructible_v<_Sent, _Sent&>)
        requires (_Kind == subrange_kind::sized)
      : _M_begin(std::move(__i)), _M_end(__s)
      {
        if constexpr (_S_store_size)
          _M_size._M_size = __n;
      }

_M_size._M_size in the function body is already const.

^ permalink raw reply	[flat|nested] 13+ messages in thread

* [Bug libstdc++/111948] subrange modifies a const size object
  2023-10-24  5:51 [Bug libstdc++/111948] New: subrange modifies a const size object hewillk at gmail dot com
  2023-10-24  5:55 ` [Bug libstdc++/111948] " hewillk at gmail dot com
@ 2023-10-24 16:15 ` redi at gcc dot gnu.org
  2023-10-24 16:20 ` redi at gcc dot gnu.org
                   ` (9 subsequent siblings)
  11 siblings, 0 replies; 13+ messages in thread
From: redi at gcc dot gnu.org @ 2023-10-24 16:15 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111948

--- Comment #2 from Jonathan Wakely <redi at gcc dot gnu.org> ---
(In reply to 康桓瑋 from comment #1)
> _M_size._M_size in the function body is already const.

It shouldn't be. Is that a compiler bug?

Clang compiles the same libstdc++ code without problems.

^ permalink raw reply	[flat|nested] 13+ messages in thread

* [Bug libstdc++/111948] subrange modifies a const size object
  2023-10-24  5:51 [Bug libstdc++/111948] New: subrange modifies a const size object hewillk at gmail dot com
  2023-10-24  5:55 ` [Bug libstdc++/111948] " hewillk at gmail dot com
  2023-10-24 16:15 ` redi at gcc dot gnu.org
@ 2023-10-24 16:20 ` redi at gcc dot gnu.org
  2023-10-24 16:30 ` redi at gcc dot gnu.org
                   ` (8 subsequent siblings)
  11 siblings, 0 replies; 13+ messages in thread
From: redi at gcc dot gnu.org @ 2023-10-24 16:20 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111948

--- Comment #3 from Jonathan Wakely <redi at gcc dot gnu.org> ---
This compiles OK:

namespace std::ranges
{
  template<class T, bool Sized = true>
    class subrange
    {
    private:
      template<typename, bool = Sized>
        struct _Size
        { };

      template<typename U>
        struct _Size<U, true>
        { U _M_size; };

      [[no_unique_address]] _Size<unsigned> _M_size = {};

    public:
      constexpr
      subrange(const T*, unsigned __n)
      {
        if constexpr (Sized)
          _M_size._M_size = __n;
      }

      constexpr unsigned size() const
      {
        if constexpr (Sized)
          return _M_size._M_size;
        else
          return 0;
      }
    };

  template<class T>
  constexpr int distance(subrange<T, true> const& s)
  { return s.size(); }

} // std::ranges

constexpr int i = 0;
constexpr auto r = std::ranges::subrange(&i, 1);
static_assert(std::ranges::distance(r));

^ permalink raw reply	[flat|nested] 13+ messages in thread

* [Bug libstdc++/111948] subrange modifies a const size object
  2023-10-24  5:51 [Bug libstdc++/111948] New: subrange modifies a const size object hewillk at gmail dot com
                   ` (2 preceding siblings ...)
  2023-10-24 16:20 ` redi at gcc dot gnu.org
@ 2023-10-24 16:30 ` redi at gcc dot gnu.org
  2023-10-24 20:37 ` hewillk at gmail dot com
                   ` (7 subsequent siblings)
  11 siblings, 0 replies; 13+ messages in thread
From: redi at gcc dot gnu.org @ 2023-10-24 16:30 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111948

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever confirmed|0                           |1
   Last reconfirmed|                            |2023-10-24

--- Comment #4 from Jonathan Wakely <redi at gcc dot gnu.org> ---
This patch makes the testcase compile, but I still think it's a compiler bug
not a library bug:

--- a/libstdc++-v3/include/bits/ranges_util.h
+++ b/libstdc++-v3/include/bits/ranges_util.h
@@ -267,13 +267,21 @@ namespace ranges
       using __size_type
        = __detail::__make_unsigned_like_t<iter_difference_t<_It>>;

-      template<typename, bool = _S_store_size>
+      template<typename _Tp, bool = _S_store_size>
        struct _Size
-       { };
+       {
+         [[__gnu__::__always_inline__]]
+         constexpr _Size(_Tp) { }
+       };

       template<typename _Tp>
        struct _Size<_Tp, true>
-       { _Tp _M_size; };
+       {
+         [[__gnu__::__always_inline__]]
+         constexpr _Size(_Tp __s) : _M_size(__s) { }
+
+         _Tp _M_size;
+       };

       [[no_unique_address]] _Size<__size_type> _M_size = {};

@@ -294,11 +302,8 @@ namespace ranges
       noexcept(is_nothrow_constructible_v<_It, decltype(__i)>
               && is_nothrow_constructible_v<_Sent, _Sent&>)
        requires (_Kind == subrange_kind::sized)
-      : _M_begin(std::move(__i)), _M_end(__s)
-      {
-       if constexpr (_S_store_size)
-         _M_size._M_size = __n;
-      }
+      : _M_begin(std::move(__i)), _M_end(__s), _M_size(__n)
+      { }

       template<__detail::__different_from<subrange> _Rng>
        requires borrowed_range<_Rng>

^ permalink raw reply	[flat|nested] 13+ messages in thread

* [Bug libstdc++/111948] subrange modifies a const size object
  2023-10-24  5:51 [Bug libstdc++/111948] New: subrange modifies a const size object hewillk at gmail dot com
                   ` (3 preceding siblings ...)
  2023-10-24 16:30 ` redi at gcc dot gnu.org
@ 2023-10-24 20:37 ` hewillk at gmail dot com
  2023-11-08 11:35 ` de34 at live dot cn
                   ` (6 subsequent siblings)
  11 siblings, 0 replies; 13+ messages in thread
From: hewillk at gmail dot com @ 2023-10-24 20:37 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111948

--- Comment #5 from 康桓瑋 <hewillk at gmail dot com> ---
It shouldn't be. Is that a compiler bug?

Clang compiles the same libstdc++ code without problems.(In reply to Jonathan
Wakely from comment #2)
> (In reply to 康桓瑋 from comment #1)
> > _M_size._M_size in the function body is already const.
> 
> It shouldn't be. Is that a compiler bug?
> 
> Clang compiles the same libstdc++ code without problems.

Sorry, I didn't delve into whether this was a bug in libstdc++ or the compiler.
At the moment it seems this is a compiler problem.

https://godbolt.org/z/ronn4exhG

^ permalink raw reply	[flat|nested] 13+ messages in thread

* [Bug libstdc++/111948] subrange modifies a const size object
  2023-10-24  5:51 [Bug libstdc++/111948] New: subrange modifies a const size object hewillk at gmail dot com
                   ` (4 preceding siblings ...)
  2023-10-24 20:37 ` hewillk at gmail dot com
@ 2023-11-08 11:35 ` de34 at live dot cn
  2023-11-08 12:01 ` redi at gcc dot gnu.org
                   ` (5 subsequent siblings)
  11 siblings, 0 replies; 13+ messages in thread
From: de34 at live dot cn @ 2023-11-08 11:35 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111948

Jiang An <de34 at live dot cn> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |de34 at live dot cn

--- Comment #6 from Jiang An <de34 at live dot cn> ---
This is certainly a compiler bug. I've reduced it and reported Bug 112439.

^ permalink raw reply	[flat|nested] 13+ messages in thread

* [Bug libstdc++/111948] subrange modifies a const size object
  2023-10-24  5:51 [Bug libstdc++/111948] New: subrange modifies a const size object hewillk at gmail dot com
                   ` (5 preceding siblings ...)
  2023-11-08 11:35 ` de34 at live dot cn
@ 2023-11-08 12:01 ` redi at gcc dot gnu.org
  2023-11-23 14:40 ` redi at gcc dot gnu.org
                   ` (4 subsequent siblings)
  11 siblings, 0 replies; 13+ messages in thread
From: redi at gcc dot gnu.org @ 2023-11-08 12:01 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111948

--- Comment #7 from Jonathan Wakely <redi at gcc dot gnu.org> ---
Thanks! I briefly tried but failed to reproduce it in a reduced example.

^ permalink raw reply	[flat|nested] 13+ messages in thread

* [Bug libstdc++/111948] subrange modifies a const size object
  2023-10-24  5:51 [Bug libstdc++/111948] New: subrange modifies a const size object hewillk at gmail dot com
                   ` (6 preceding siblings ...)
  2023-11-08 12:01 ` redi at gcc dot gnu.org
@ 2023-11-23 14:40 ` redi at gcc dot gnu.org
  2023-12-05 23:35 ` cvs-commit at gcc dot gnu.org
                   ` (3 subsequent siblings)
  11 siblings, 0 replies; 13+ messages in thread
From: redi at gcc dot gnu.org @ 2023-11-23 14:40 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111948

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Depends on|                            |112439
           Assignee|unassigned at gcc dot gnu.org      |redi at gcc dot gnu.org
             Status|NEW                         |ASSIGNED
           See Also|https://gcc.gnu.org/bugzill |
                   |a/show_bug.cgi?id=112439    |
   Target Milestone|---                         |13.3

--- Comment #8 from Jonathan Wakely <redi at gcc dot gnu.org> ---
This hasn't been fixed yet, so I'll add the library workaround.


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112439
[Bug 112439] [13/14 Regression] Modification of a member overlapping with a
[[no_unique_address]] member in the constructor is incorrectly rejected in
constant evaluation

^ permalink raw reply	[flat|nested] 13+ messages in thread

* [Bug libstdc++/111948] subrange modifies a const size object
  2023-10-24  5:51 [Bug libstdc++/111948] New: subrange modifies a const size object hewillk at gmail dot com
                   ` (7 preceding siblings ...)
  2023-11-23 14:40 ` redi at gcc dot gnu.org
@ 2023-12-05 23:35 ` cvs-commit at gcc dot gnu.org
  2023-12-06 14:44 ` cvs-commit at gcc dot gnu.org
                   ` (2 subsequent siblings)
  11 siblings, 0 replies; 13+ messages in thread
From: cvs-commit at gcc dot gnu.org @ 2023-12-05 23:35 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111948

--- Comment #9 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Jonathan Wakely <redi@gcc.gnu.org>:

https://gcc.gnu.org/g:08448dc146b6dd32383d64ab491a594d41f62aaa

commit r14-6200-g08448dc146b6dd32383d64ab491a594d41f62aaa
Author: Jonathan Wakely <jwakely@redhat.com>
Date:   Tue Oct 24 20:15:12 2023 +0100

    libstdc++: Add workaround to std::ranges::subrange [PR111948]

    libstdc++-v3/ChangeLog:

            PR libstdc++/111948
            * include/bits/ranges_util.h (subrange): Add constructor to
            _Size to aoid setting member in constructor.
            * testsuite/std/ranges/subrange/111948.cc: New test.

^ permalink raw reply	[flat|nested] 13+ messages in thread

* [Bug libstdc++/111948] subrange modifies a const size object
  2023-10-24  5:51 [Bug libstdc++/111948] New: subrange modifies a const size object hewillk at gmail dot com
                   ` (8 preceding siblings ...)
  2023-12-05 23:35 ` cvs-commit at gcc dot gnu.org
@ 2023-12-06 14:44 ` cvs-commit at gcc dot gnu.org
  2023-12-06 14:46 ` redi at gcc dot gnu.org
  2024-02-02 16:12 ` jason at gcc dot gnu.org
  11 siblings, 0 replies; 13+ messages in thread
From: cvs-commit at gcc dot gnu.org @ 2023-12-06 14:44 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111948

--- Comment #10 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-13 branch has been updated by Jonathan Wakely
<redi@gcc.gnu.org>:

https://gcc.gnu.org/g:be26992b3115b95189ac02fcf25902cb5430c0e1

commit r13-8133-gbe26992b3115b95189ac02fcf25902cb5430c0e1
Author: Jonathan Wakely <jwakely@redhat.com>
Date:   Tue Oct 24 20:15:12 2023 +0100

    libstdc++: Add workaround to std::ranges::subrange [PR111948]

    libstdc++-v3/ChangeLog:

            PR libstdc++/111948
            * include/bits/ranges_util.h (subrange): Add constructor to
            _Size to avoid setting member in constructor.
            * testsuite/std/ranges/subrange/111948.cc: New test.

    (cherry picked from commit 08448dc146b6dd32383d64ab491a594d41f62aaa)

^ permalink raw reply	[flat|nested] 13+ messages in thread

* [Bug libstdc++/111948] subrange modifies a const size object
  2023-10-24  5:51 [Bug libstdc++/111948] New: subrange modifies a const size object hewillk at gmail dot com
                   ` (9 preceding siblings ...)
  2023-12-06 14:44 ` cvs-commit at gcc dot gnu.org
@ 2023-12-06 14:46 ` redi at gcc dot gnu.org
  2024-02-02 16:12 ` jason at gcc dot gnu.org
  11 siblings, 0 replies; 13+ messages in thread
From: redi at gcc dot gnu.org @ 2023-12-06 14:46 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111948

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

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

--- Comment #11 from Jonathan Wakely <redi at gcc dot gnu.org> ---
Fixed for 13.3, thanks for the report.

^ permalink raw reply	[flat|nested] 13+ messages in thread

* [Bug libstdc++/111948] subrange modifies a const size object
  2023-10-24  5:51 [Bug libstdc++/111948] New: subrange modifies a const size object hewillk at gmail dot com
                   ` (10 preceding siblings ...)
  2023-12-06 14:46 ` redi at gcc dot gnu.org
@ 2024-02-02 16:12 ` jason at gcc dot gnu.org
  11 siblings, 0 replies; 13+ messages in thread
From: jason at gcc dot gnu.org @ 2024-02-02 16:12 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111948
Bug 111948 depends on bug 112439, which changed state.

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

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

^ permalink raw reply	[flat|nested] 13+ messages in thread

end of thread, other threads:[~2024-02-02 16:12 UTC | newest]

Thread overview: 13+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-10-24  5:51 [Bug libstdc++/111948] New: subrange modifies a const size object hewillk at gmail dot com
2023-10-24  5:55 ` [Bug libstdc++/111948] " hewillk at gmail dot com
2023-10-24 16:15 ` redi at gcc dot gnu.org
2023-10-24 16:20 ` redi at gcc dot gnu.org
2023-10-24 16:30 ` redi at gcc dot gnu.org
2023-10-24 20:37 ` hewillk at gmail dot com
2023-11-08 11:35 ` de34 at live dot cn
2023-11-08 12:01 ` redi at gcc dot gnu.org
2023-11-23 14:40 ` redi at gcc dot gnu.org
2023-12-05 23:35 ` cvs-commit at gcc dot gnu.org
2023-12-06 14:44 ` cvs-commit at gcc dot gnu.org
2023-12-06 14:46 ` redi at gcc dot gnu.org
2024-02-02 16:12 ` jason at gcc dot gnu.org

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).