[Bug c++/61556] New: [c++11][regression] ‘*(const ValueType*)this’ is not a constant expression with valid code

[Bug c++/61556] New: [c++11][regression] ‘*(const ValueType*)this’ is not a constant expression with valid code

[l.jirkovsky at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61556

            Bug ID: 61556
           Summary: [c++11][regression] ‘*(const ValueType*)this’ is not a
                    constant expression with valid code
           Product: gcc
           Version: 4.9.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: l.jirkovsky at gmail dot com

When used in a certain way, a template class cannot be instantiated in another
template class using a static constexpr. I tracked the regression down to the
following commit:

commit 18619da58c77461642c36cee9f0463c968877f75
Author: jason <jason@138bc75d-0d04-0410-961f-82ee72b054a4>
Date:   Mon Apr 1 19:05:12 2013 +0000

        * call.c (add_function_candidate): Take the address of 'this' here.
        (build_over_call): And here.
        (build_new_method_call_1, build_op_call_1): Not here.
        (build_user_type_conversion_1): Or here.
        (add_candidates): Adjust.

Reverting this commit fixes the problem on the current gcc.

The problematic code that should be compilable is attached after the "CODE"
section.

When compiled with gcc 4.9.0, the
ValueTypeInfo<ValueTypeEnum::doubleval> v; in the FillFunctor
constructor gives an error:

    bug2.cpp: In constructor ‘FillFunctor<Format>::FillFunctor()’:
    bug2.cpp:24:41:   in constexpr expansion of ‘const
ValueType{1}->ValueType::operator int()’
    bug2.cpp:24:41: error: ‘*(const ValueType*)this’ is not a constant
expression
       ValueTypeInfo<ValueTypeEnum::doubleval> v;
            ^
    bug2.cpp:24:41: note: in template argument for type ‘int’
    bug2.cpp:24:44: error: invalid type in declaration before ‘;’ token
       ValueTypeInfo<ValueTypeEnum::doubleval> v;


However, the same expression compiles just fine if its used in a
function, such as main(). Removing m_ID makes the code compile, too.
Also, the compilation works after replacing:

  ValueTypeInfo<ValueTypeEnum::doubleval> v;

with a:

  constexpr ValueType doubleval = ValueType(1);
  ValueTypeInfo<doubleval> v;

CODE
#####

class ValueType {
public:
    constexpr operator int() const {return m_ID;};
    constexpr ValueType(const int v)
        : m_ID(v) {}
private:
    int m_ID;
};

class ValueTypeEnum {
public:
    static constexpr ValueType doubleval = ValueType(1);
};

template <int format>
class ValueTypeInfo {
};

template <typename Format>
class FillFunctor {
public:
    FillFunctor() {
        ValueTypeInfo<ValueTypeEnum::doubleval> v;
    }
};

int main() {
    ValueTypeInfo<ValueTypeEnum::doubleval> v;
}
>From gcc-bugs-return-454524-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Thu Jun 19 00:45:18 2014
Return-Path: <gcc-bugs-return-454524-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 12505 invoked by alias); 19 Jun 2014 00:45:17 -0000
Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-bugs.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-help@gcc.gnu.org>
Sender: gcc-bugs-owner@gcc.gnu.org
Delivered-To: mailing list gcc-bugs@gcc.gnu.org
Received: (qmail 12458 invoked by uid 48); 19 Jun 2014 00:45:11 -0000
From: "kumba at gentoo dot org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug c++/61538] g++ compiled binary, linked to glibc libpthread, hangs on SGI MIPS R1x000 systems on Linux
Date: Thu, 19 Jun 2014 00:45:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: c++
X-Bugzilla-Version: 4.9.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: major
X-Bugzilla-Who: kumba at gentoo dot org
X-Bugzilla-Status: UNCONFIRMED
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields:
Message-ID: <bug-61538-4-2FxSxnVLMT@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-61538-4@http.gcc.gnu.org/bugzilla/>
References: <bug-61538-4@http.gcc.gnu.org/bugzilla/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
MIME-Version: 1.0
X-SW-Source: 2014-06/txt/msg01606.txt.bz2
Content-length: 2114

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61538

--- Comment #4 from Joshua Kinard <kumba at gentoo dot org> ---
It looks like the bug might be somewhere in __cxa_guard_acquire() in
libstdc++-v3/lubsupc++/guard.cc, as that references glibc and futexes.  strace
indicates that the same syscall was invoked twice in a row -- could this be a
double-locking bug?

This is what I traced out in gdb:

   ¦0x77ed91cc <__cxa_guard_acquire+336>    sw     zero,32(sp)
B+>¦0x77ed91d0 <__cxa_guard_acquire+340>    b      0x77ed9144
<__cxa_guard_acquire+200>
    |
    |->¦0x77ed9144 <__cxa_guard_acquire+200>    lw     t9,-28620(gp)
       ¦0x77ed9148 <__cxa_guard_acquire+204>    li     a0,4238
       ¦0x77ed914c <__cxa_guard_acquire+208>    move   a1,s0
       ¦0x77ed9150 <__cxa_guard_acquire+212>    move   a2,zero
       ¦0x77ed9154 <__cxa_guard_acquire+216>    lw     a3,32(sp)
       ¦0x77ed9158 <__cxa_guard_acquire+220>    jalr   t9
        |
        |->¦0x77d50850 <syscall>    lui    gp,0x9
           ¦0x77d50854 <syscall+4>  addiu  gp,gp,-2624
           ¦0x77d50858 <syscall+8>  addu   gp,gp,t9
           ¦0x77d5085c <syscall+12> li     v0,4000
           ¦0x77d50860 <syscall+16> syscall
               <HANGS HERE>
           ¦0x77d50864 <syscall+20> bnez   a3,0x77d50840
           ¦0x77d50868 <syscall+24> nop
           ¦0x77d5086c <syscall+28> jr     ra
           ¦0x77d50870 <syscall+32> nop
           ¦0x77d50874 <syscall+36> nop
           ¦0x77d50878              nop
           ¦0x77d5087c              nop

I can see the first futex syscall (li a0,4238), and I think it looks like
inside that syscall, it's doing some loads and adds, then makes a "generic"
syscall (#4000), probably passing the computed 0x108e value as the first
argument, which would translate into another futex syscall, which jives with
what strace says.  Is taking a futex inside of a futex a good thing?  It's
obvious that something with the R1x000 CPU is coming into play as well, but I
don't know what exactly.
>From gcc-bugs-return-454525-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Thu Jun 19 01:08:38 2014
Return-Path: <gcc-bugs-return-454525-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 7714 invoked by alias); 19 Jun 2014 01:08:38 -0000
Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-bugs.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-help@gcc.gnu.org>
Sender: gcc-bugs-owner@gcc.gnu.org
Delivered-To: mailing list gcc-bugs@gcc.gnu.org
Received: (qmail 7667 invoked by uid 48); 19 Jun 2014 01:08:31 -0000
From: "pinskia at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug c++/61538] g++ compiled binary, linked to glibc libpthread, hangs on SGI MIPS R1x000 systems on Linux
Date: Thu, 19 Jun 2014 01:08:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: c++
X-Bugzilla-Version: 4.9.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: major
X-Bugzilla-Who: pinskia at gcc dot gnu.org
X-Bugzilla-Status: UNCONFIRMED
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields:
Message-ID: <bug-61538-4-X58FhhYT7u@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-61538-4@http.gcc.gnu.org/bugzilla/>
References: <bug-61538-4@http.gcc.gnu.org/bugzilla/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
MIME-Version: 1.0
X-SW-Source: 2014-06/txt/msg01607.txt.bz2
Content-length: 175

https://gcc.gnu.org/bugzilla/show_bug.cgi?ida538

--- Comment #5 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
https://gcc.gnu.org/bugzilla/show_bug.cgi?idR839#c12

[Bug c++/61556] [c++11][4.9/4.10 Regression] ‘*(const ValueType*)this’ is not a constant expression with valid code

[jason at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61556

--- Comment #1 from Jason Merrill <jason at gcc dot gnu.org> ---
Author: jason
Date: Fri Jun 20 18:31:53 2014
New Revision: 211853

URL: https://gcc.gnu.org/viewcvs?rev=211853&root=gcc&view=rev
Log:
    PR c++/61556
    * call.c (build_over_call): Call build_this in template path.

Added:
    trunk/gcc/testsuite/g++.dg/cpp0x/constexpr-template7.C
Modified:
    trunk/gcc/cp/ChangeLog
    trunk/gcc/cp/call.c

[Bug c++/61556] [c++11][4.9/4.10 Regression] ‘*(const ValueType*)this’ is not a constant expression with valid code

[jason at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61556

--- Comment #2 from Jason Merrill <jason at gcc dot gnu.org> ---
Author: jason
Date: Fri Jun 20 18:34:37 2014
New Revision: 211855

URL: https://gcc.gnu.org/viewcvs?rev=211855&root=gcc&view=rev
Log:
    PR c++/61556
    * call.c (build_over_call): Call build_this in template path.

Added:
    branches/gcc-4_9-branch/gcc/testsuite/g++.dg/cpp0x/constexpr-template7.C
Modified:
    branches/gcc-4_9-branch/gcc/cp/ChangeLog
    branches/gcc-4_9-branch/gcc/cp/call.c

[Bug c++/61556] [c++11][4.9/4.10 Regression] ‘*(const ValueType*)this’ is not a constant expression with valid code

[paolo.carlini at oracle dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61556

Paolo Carlini <paolo.carlini at oracle dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|---                         |FIXED
   Target Milestone|---                         |4.9.1

--- Comment #3 from Paolo Carlini <paolo.carlini at oracle dot com> ---
Thus fixed.