public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug c++/108309] New: ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136
@ 2023-01-06  3:09 tim at klingt dot org
  2023-01-06  3:09 ` [Bug c++/108309] " tim at klingt dot org
                   ` (7 more replies)
  0 siblings, 8 replies; 9+ messages in thread
From: tim at klingt dot org @ 2023-01-06  3:09 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108309

            Bug ID: 108309
           Summary: ICE in in cxx_eval_component_reference, at
                    cp/constexpr.cc:4136
           Product: gcc
           Version: 12.2.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: tim at klingt dot org
  Target Milestone: ---

```
template <typename _Tp> constexpr _Tp forward(_Tp &__t) {
  return static_cast<_Tp &&>(__t);
}
struct _Enable_copy_move {};
struct optional;
template <int _Nm> struct __array_traits { typedef optional _Type[_Nm]; };
template <int _Nm> struct array { __array_traits<_Nm>::_Type _M_elems; };
struct _Optional_payload_base {
  union {
    int _M_value;
  };
};
struct _Optional_base {
  constexpr _Optional_base(...) : _M_payload() {}
  constexpr _Optional_base(_Optional_base &&) {}
  _Optional_payload_base _M_payload;
};
struct optional : _Optional_base, _Enable_copy_move {
  using _Base = _Optional_base;
  constexpr optional() {}
  template <typename _Up> constexpr optional(_Up __t) : _Base(forward(__t)) {}
} page {
  array<8> {}
}
```

gives me:

```
➜  scratch /usr/bin/g++-12 -std=gnu++20 testcase.ii
testcase.ii:24:1:   in ‘constexpr’ expansion of
‘page.optional::optional<array<8> >(array<8>{__array_traits<8>::_Type()})’
testcase.ii:24:1:   in ‘constexpr’ expansion of ‘forward(_Tp&) [with _Tp =
array<8>]()’
testcase.ii:24:1:   in ‘constexpr’ expansion of ‘array<8>((*
&((array<8>&&)__t)))’
testcase.ii:24:1:   in ‘constexpr’ expansion of
‘optional(<anonymous>.array<8>::_M_elems[0])’
testcase.ii:24:1: internal compiler error: in cxx_eval_component_reference, at
cp/constexpr.cc:4136
   24 | }
      | ^
0x7fb4eb429d8f __libc_start_call_main
        ../sysdeps/nptl/libc_start_call_main.h:58
0x7fb4eb429e3f __libc_start_main_impl
        ../csu/libc-start.c:392
Please submit a full bug report, with preprocessed source (by using
-freport-bug).
Please include the complete backtrace with any bug report.
See <file:///usr/share/doc/gcc-12/README.Bugs> for instructions.
```

with
```
g++-12 (Ubuntu 12.2.0-3ubuntu1~22.04) 12.2.0
Copyright (C) 2022 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
```

unfortunately cvise didn't reduce the test case to valid c++ code (afaict
there's only a missing semicolon, though). with a trailing semicolon, gcc-11
accepts the code, but clang++-15 doesn't (not sure if it's related, but i
thought it's worth mentioning:
```
testcase.reduced.cc:21:63: error: cannot pass object of non-trivial type
'array<8>' through variadic constructor; call will abort at runtime
[-Wnon-pod-varargs]
  template <typename _Up> constexpr optional(_Up __t) : _Base(forward(__t)) {}
                                                              ^
testcase.reduced.cc:22:3: note: in instantiation of function template
specialization 'optional::optional<array<8>>' requested here
} page {
  ^
1 error generated.
```

^ permalink raw reply	[flat|nested] 9+ messages in thread

* [Bug c++/108309] ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136
  2023-01-06  3:09 [Bug c++/108309] New: ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136 tim at klingt dot org
@ 2023-01-06  3:09 ` tim at klingt dot org
  2023-01-06  6:22 ` [Bug c++/108309] [12 Regression] " pinskia at gcc dot gnu.org
                   ` (6 subsequent siblings)
  7 siblings, 0 replies; 9+ messages in thread
From: tim at klingt dot org @ 2023-01-06  3:09 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108309

--- Comment #1 from tim blechmann <tim at klingt dot org> ---
Created attachment 54200
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=54200&action=edit
reduced test case

^ permalink raw reply	[flat|nested] 9+ messages in thread

* [Bug c++/108309] [12 Regression] ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136
  2023-01-06  3:09 [Bug c++/108309] New: ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136 tim at klingt dot org
  2023-01-06  3:09 ` [Bug c++/108309] " tim at klingt dot org
@ 2023-01-06  6:22 ` pinskia at gcc dot gnu.org
  2023-01-06  6:31 ` tim at klingt dot org
                   ` (5 subsequent siblings)
  7 siblings, 0 replies; 9+ messages in thread
From: pinskia at gcc dot gnu.org @ 2023-01-06  6:22 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108309

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |12.3
           Keywords|                            |ice-on-valid-code,
                   |                            |needs-bisection
      Known to fail|                            |12.2.0
            Summary|ICE in in                   |[12 Regression] ICE in in
                   |cxx_eval_component_referenc |cxx_eval_component_referenc
                   |e, at cp/constexpr.cc:4136  |e, at cp/constexpr.cc:4136
      Known to work|                            |11.3.0
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2023-01-06

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Reduced further:
```
struct _Enable_copy_move {};
struct h {
  union {
    int _M_value;
  };
};
struct m {
  constexpr m(...) : _M_payload() {}
  constexpr m(const m &) {}
  h _M_payload;
};
struct optional : m, _Enable_copy_move {
  constexpr optional() {}
  template <typename _Up> constexpr optional(_Up __t) : m(__t) {}
};
struct array { optional t[1]; };

optional page {
  array {}
};

```

>clang++-15 doesn't (not sure if it's related, but i thought it's worth mentioning:

That is a conditionally supported construct in C++. clang does not support it
but GCC does. 

GCC does warn with -Wconditionally-supported :

<source>: In instantiation of 'constexpr optional::optional(_Up) [with _Up =
array]':
<source>:20:1:   required from here
<source>:14:62: warning: passing objects of non-trivially-copyable type 'struct
array' through '...' is conditionally supported [-Wconditionally-supported]
   14 |   template <typename _Up> constexpr optional(_Up __t) : m(__t) {}
      |                                                              ^


Anyways this is fixed on the trunk; though since it is a GCC 12.x regression it
will/should be fixed on the GCC 12 branch too once someone figures out which
patch fixed it.

^ permalink raw reply	[flat|nested] 9+ messages in thread

* [Bug c++/108309] [12 Regression] ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136
  2023-01-06  3:09 [Bug c++/108309] New: ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136 tim at klingt dot org
  2023-01-06  3:09 ` [Bug c++/108309] " tim at klingt dot org
  2023-01-06  6:22 ` [Bug c++/108309] [12 Regression] " pinskia at gcc dot gnu.org
@ 2023-01-06  6:31 ` tim at klingt dot org
  2023-01-06  6:33 ` pinskia at gcc dot gnu.org
                   ` (4 subsequent siblings)
  7 siblings, 0 replies; 9+ messages in thread
From: tim at klingt dot org @ 2023-01-06  6:31 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108309

--- Comment #3 from tim blechmann <tim at klingt dot org> ---
> That is a conditionally supported construct in C++. clang does not support it but GCC does. 
> 
> GCC does warn with -Wconditionally-supported

interesting. i may have to reduce it manually rather than going through cvise:
the real-world code builds with clang and gcc doesn't seem to emit a warning
with -Wconditionally-supported

^ permalink raw reply	[flat|nested] 9+ messages in thread

* [Bug c++/108309] [12 Regression] ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136
  2023-01-06  3:09 [Bug c++/108309] New: ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136 tim at klingt dot org
                   ` (2 preceding siblings ...)
  2023-01-06  6:31 ` tim at klingt dot org
@ 2023-01-06  6:33 ` pinskia at gcc dot gnu.org
  2023-01-06 11:30 ` marxin at gcc dot gnu.org
                   ` (3 subsequent siblings)
  7 siblings, 0 replies; 9+ messages in thread
From: pinskia at gcc dot gnu.org @ 2023-01-06  6:33 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108309

--- Comment #4 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to tim blechmann from comment #3)
> > That is a conditionally supported construct in C++. clang does not support it but GCC does. 
> > 
> > GCC does warn with -Wconditionally-supported
> 
> interesting. i may have to reduce it manually rather than going through
> cvise: the real-world code builds with clang and gcc doesn't seem to emit a
> warning with -Wconditionally-supported

Or you could add -Werror=conditionally-supported as an option and reduce it
that way ...

^ permalink raw reply	[flat|nested] 9+ messages in thread

* [Bug c++/108309] [12 Regression] ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136
  2023-01-06  3:09 [Bug c++/108309] New: ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136 tim at klingt dot org
                   ` (3 preceding siblings ...)
  2023-01-06  6:33 ` pinskia at gcc dot gnu.org
@ 2023-01-06 11:30 ` marxin at gcc dot gnu.org
  2023-01-06 17:25 ` pinskia at gcc dot gnu.org
                   ` (2 subsequent siblings)
  7 siblings, 0 replies; 9+ messages in thread
From: marxin at gcc dot gnu.org @ 2023-01-06 11:30 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108309

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|needs-bisection             |
                 CC|                            |jason at gcc dot gnu.org,
                   |                            |marxin at gcc dot gnu.org,
                   |                            |mpolacek at gcc dot gnu.org

--- Comment #5 from Martin Liška <marxin at gcc dot gnu.org> ---
Fixed on master with r13-963-gd19b4342c19e5a7f and it started with
r12-2304-ga42f8120442cf3ba.

^ permalink raw reply	[flat|nested] 9+ messages in thread

* [Bug c++/108309] [12 Regression] ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136
  2023-01-06  3:09 [Bug c++/108309] New: ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136 tim at klingt dot org
                   ` (4 preceding siblings ...)
  2023-01-06 11:30 ` marxin at gcc dot gnu.org
@ 2023-01-06 17:25 ` pinskia at gcc dot gnu.org
  2023-04-27 12:06 ` rguenth at gcc dot gnu.org
  2023-05-08 12:26 ` rguenth at gcc dot gnu.org
  7 siblings, 0 replies; 9+ messages in thread
From: pinskia at gcc dot gnu.org @ 2023-01-06 17:25 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108309

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           See Also|                            |https://gcc.gnu.org/bugzill
                   |                            |a/show_bug.cgi?id=105795,
                   |                            |https://gcc.gnu.org/bugzill
                   |                            |a/show_bug.cgi?id=101371

--- Comment #6 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Martin Liška from comment #5)
> Fixed on master with r13-963-gd19b4342c19e5a7f

I didn't mention it when I produced the reduced testcase but removing the empty
base class does cause the ICE to go away so this looks like a real fix ...

^ permalink raw reply	[flat|nested] 9+ messages in thread

* [Bug c++/108309] [12 Regression] ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136
  2023-01-06  3:09 [Bug c++/108309] New: ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136 tim at klingt dot org
                   ` (5 preceding siblings ...)
  2023-01-06 17:25 ` pinskia at gcc dot gnu.org
@ 2023-04-27 12:06 ` rguenth at gcc dot gnu.org
  2023-05-08 12:26 ` rguenth at gcc dot gnu.org
  7 siblings, 0 replies; 9+ messages in thread
From: rguenth at gcc dot gnu.org @ 2023-04-27 12:06 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108309

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P2

^ permalink raw reply	[flat|nested] 9+ messages in thread

* [Bug c++/108309] [12 Regression] ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136
  2023-01-06  3:09 [Bug c++/108309] New: ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136 tim at klingt dot org
                   ` (6 preceding siblings ...)
  2023-04-27 12:06 ` rguenth at gcc dot gnu.org
@ 2023-05-08 12:26 ` rguenth at gcc dot gnu.org
  7 siblings, 0 replies; 9+ messages in thread
From: rguenth at gcc dot gnu.org @ 2023-05-08 12:26 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108309

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

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

--- Comment #7 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 12.3 is being released, retargeting bugs to GCC 12.4.

^ permalink raw reply	[flat|nested] 9+ messages in thread

end of thread, other threads:[~2023-05-08 12:26 UTC | newest]

Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-01-06  3:09 [Bug c++/108309] New: ICE in in cxx_eval_component_reference, at cp/constexpr.cc:4136 tim at klingt dot org
2023-01-06  3:09 ` [Bug c++/108309] " tim at klingt dot org
2023-01-06  6:22 ` [Bug c++/108309] [12 Regression] " pinskia at gcc dot gnu.org
2023-01-06  6:31 ` tim at klingt dot org
2023-01-06  6:33 ` pinskia at gcc dot gnu.org
2023-01-06 11:30 ` marxin at gcc dot gnu.org
2023-01-06 17:25 ` pinskia at gcc dot gnu.org
2023-04-27 12:06 ` rguenth at gcc dot gnu.org
2023-05-08 12:26 ` rguenth 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).