[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[hubicka at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

Jan Hubicka <hubicka at gcc dot gnu.org> changed:

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

--- Comment #18 from Jan Hubicka <hubicka at gcc dot gnu.org> 2010-11-10 22:12:42 UTC ---
Martin,
is this solved now?

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[owner at bugs dot debian.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

--- Comment #19 from owner at bugs dot debian.org 2010-11-10 22:24:15 UTC ---
Thank you for the additional information you have supplied regarding
this Bug report.

This is an automatically generated reply to let you know your message
has been received.

Your message is being forwarded to the package maintainers and other
interested parties for their attention; they will reply in due course.

Your message has been sent to the package maintainer(s):
 Debian GCC Maintainers <debian-gcc@lists.debian.org>

If you wish to submit further information on this problem, please
send it to 173513@bugs.debian.org.

Please do not send mail to owner@bugs.debian.org unless you wish
to report a problem with the Bug-tracking system.

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[owner at bugs dot debian.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

--- Comment #20 from owner at bugs dot debian.org 2010-11-10 22:27:14 UTC ---
Thank you for the additional information you have supplied regarding
this Bug report.

This is an automatically generated reply to let you know your message
has been received.

Your message is being forwarded to the package maintainers and other
interested parties for their attention; they will reply in due course.

Your message has been sent to the package maintainer(s):
 Debian GCC Maintainers <debian-gcc@lists.debian.org>

If you wish to submit further information on this problem, please
send it to 173513@bugs.debian.org.

Please do not send mail to owner@bugs.debian.org unless you wish
to report a problem with the Bug-tracking system.

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[owner at bugs dot debian.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

--- Comment #21 from owner at bugs dot debian.org 2010-11-10 22:30:23 UTC ---
Thank you for the additional information you have supplied regarding
this Bug report.

This is an automatically generated reply to let you know your message
has been received.

Your message is being forwarded to the package maintainers and other
interested parties for their attention; they will reply in due course.

Your message has been sent to the package maintainer(s):
 Debian GCC Maintainers <debian-gcc@lists.debian.org>

If you wish to submit further information on this problem, please
send it to 173513@bugs.debian.org.

Please do not send mail to owner@bugs.debian.org unless you wish
to report a problem with the Bug-tracking system.

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[owner at bugs dot debian.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

--- Comment #22 from owner at bugs dot debian.org 2010-11-10 22:33:17 UTC ---
Thank you for the additional information you have supplied regarding
this Bug report.

This is an automatically generated reply to let you know your message
has been received.

Your message is being forwarded to the package maintainers and other
interested parties for their attention; they will reply in due course.

Your message has been sent to the package maintainer(s):
 Debian GCC Maintainers <debian-gcc@lists.debian.org>

If you wish to submit further information on this problem, please
send it to 173513@bugs.debian.org.

Please do not send mail to owner@bugs.debian.org unless you wish
to report a problem with the Bug-tracking system.

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[owner at bugs dot debian.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

--- Comment #23 from owner at bugs dot debian.org 2010-11-10 22:36:14 UTC ---
Thank you for the additional information you have supplied regarding
this Bug report.

This is an automatically generated reply to let you know your message
has been received.

Your message is being forwarded to the package maintainers and other
interested parties for their attention; they will reply in due course.

Your message has been sent to the package maintainer(s):
 Debian GCC Maintainers <debian-gcc@lists.debian.org>

If you wish to submit further information on this problem, please
send it to 173513@bugs.debian.org.

Please do not send mail to owner@bugs.debian.org unless you wish
to report a problem with the Bug-tracking system.

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[owner at bugs dot debian.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

--- Comment #24 from owner at bugs dot debian.org 2010-11-10 22:39:13 UTC ---
Thank you for the additional information you have supplied regarding
this Bug report.

This is an automatically generated reply to let you know your message
has been received.

Your message is being forwarded to the package maintainers and other
interested parties for their attention; they will reply in due course.

Your message has been sent to the package maintainer(s):
 Debian GCC Maintainers <debian-gcc@lists.debian.org>

If you wish to submit further information on this problem, please
send it to 173513@bugs.debian.org.

Please do not send mail to owner@bugs.debian.org unless you wish
to report a problem with the Bug-tracking system.

--- Comment #25 from owner at bugs dot debian.org 2010-11-10 22:39:14 UTC ---
Thank you for the additional information you have supplied regarding
this Bug report.

This is an automatically generated reply to let you know your message
has been received.

Your message is being forwarded to the package maintainers and other
interested parties for their attention; they will reply in due course.

Your message has been sent to the package maintainer(s):
 Debian GCC Maintainers <debian-gcc@lists.debian.org>

If you wish to submit further information on this problem, please
send it to 173513@bugs.debian.org.

Please do not send mail to owner@bugs.debian.org unless you wish
to report a problem with the Bug-tracking system.

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[owner at bugs dot debian.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

--- Comment #24 from owner at bugs dot debian.org 2010-11-10 22:39:13 UTC ---
Thank you for the additional information you have supplied regarding
this Bug report.

This is an automatically generated reply to let you know your message
has been received.

Your message is being forwarded to the package maintainers and other
interested parties for their attention; they will reply in due course.

Your message has been sent to the package maintainer(s):
 Debian GCC Maintainers <debian-gcc@lists.debian.org>

If you wish to submit further information on this problem, please
send it to 173513@bugs.debian.org.

Please do not send mail to owner@bugs.debian.org unless you wish
to report a problem with the Bug-tracking system.

--- Comment #25 from owner at bugs dot debian.org 2010-11-10 22:39:14 UTC ---
Thank you for the additional information you have supplied regarding
this Bug report.

This is an automatically generated reply to let you know your message
has been received.

Your message is being forwarded to the package maintainers and other
interested parties for their attention; they will reply in due course.

Your message has been sent to the package maintainer(s):
 Debian GCC Maintainers <debian-gcc@lists.debian.org>

If you wish to submit further information on this problem, please
send it to 173513@bugs.debian.org.

Please do not send mail to owner@bugs.debian.org unless you wish
to report a problem with the Bug-tracking system.

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[ian at airs dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

Ian Lance Taylor <ian at airs dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |ian at airs dot com

--- Comment #26 from Ian Lance Taylor <ian at airs dot com> 2010-11-10 22:49:18 UTC ---
See also http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=173513 .

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[jamborm at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

--- Comment #27 from Martin Jambor <jamborm at gcc dot gnu.org> 2010-11-11 09:45:02 UTC ---
(In reply to comment #18)
> Martin,
> is this solved now?

The last time I checked it wasn't, mainly because this is not really
an IPA problem (if it was, indirect inlining would handle it just
fine).

I believe the current plan is to make early FRE fix this, see the
discussion in PR 40874.

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

--- Comment #28 from Paolo Carlini <paolo.carlini at oracle dot com> 2010-11-14 17:11:21 UTC ---
For the record, my favorite example in the C++ library,
testsuite/performance/27_io/fmtflags_manipulators.cc, pretty simple, still
probably noticeable by many of our users, is by now perfectly optimized at -O2.
Many thanks.

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[mvanderkolff at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

--- Comment #29 from Michael van der Kolff <mvanderkolff at gmail dot com> 2013-01-16 12:38:41 UTC ---
Created attachment 29180
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=29180
testcase for member function pointer that isn't inlined

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[mvanderkolff at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

Michael van der Kolff <mvanderkolff at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mvanderkolff at gmail dot
                   |                            |com

--- Comment #30 from Michael van der Kolff <mvanderkolff at gmail dot com> 2013-01-16 12:39:24 UTC ---
The following is currently (g++ 4.7.2) inlined:
#include <iostream>

using namespace std;

class Foo
{
public:
    void Bar() const
    {
        cout << "Howdy!" << endl;
    }
};

int main()
{
    Foo x;
    auto y = [] (const Foo& z) { z.Bar(); };
    y(x);
    return 0;
}
objdump -D:
--snip --
  400700:       48 83 ec 08             sub    $0x8,%rsp
  400704:       be 0c 09 40 00          mov    $0x40090c,%esi
  400709:       bf a0 0c 60 00          mov    $0x600ca0,%edi
  40070e:       e8 cd ff ff ff          callq  4006e0
<_ZStlsISt11char_traitsIcEERSt13basic_ostreamIcT_ES5_PKc@plt>
  400713:       48 89 c7                mov    %rax,%rdi
  400716:       e8 d5 ff ff ff          callq  4006f0
<_ZSt4endlIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_@plt>
  40071b:       31 c0                   xor    %eax,%eax
  40071d:       48 83 c4 08             add    $0x8,%rsp
  400721:       c3                      retq   
  400722:       66 66 66 66 66 2e 0f    data32 data32 data32 data32 nopw
%cs:0x0(%rax,%rax,1)
--snip --
whereas any version with ptr-to-mem fn is not:
...
int main()
{
    Foo x;
    auto y = &Foo::Bar;
    (x.*y)();
    return 0;
}
objdump -D:
--snip --
0000000000400830 <main>:
  400830:       48 83 ec 18             sub    $0x18,%rsp
  400834:       48 8d 7c 24 0f          lea    0xf(%rsp),%rdi
  400839:       e8 42 01 00 00          callq  400980 <_ZNK3Foo3BarEv>
  40083e:       31 c0                   xor    %eax,%eax
  400840:       48 83 c4 18             add    $0x18,%rsp
  400844:       c3                      retq   
  400845:       66 66 2e 0f 1f 84 00    data32 nopw %cs:0x0(%rax,%rax,1)
--snip --

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[hubicka at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

--- Comment #31 from Jan Hubicka <hubicka at gcc dot gnu.org> 2013-01-16 14:20:46 UTC ---
Well, after early optimizations we get:
int main() ()
{
  struct Foo x;
  void Foo::<T392> (const struct Foo *) * iftmp.0;
  long int _3;
  long int _4;
  int (*__vtbl_ptr_type) () * _5;
  long int _6;
  sizetype _7;
  int (*__vtbl_ptr_type) () * _8;

  <bb 2>:
  _3 = (long int) Bar;
  _4 = _3 & 1;
  if (_4 == 0)
    goto <bb 4>;
  else
    goto <bb 3>;

  <bb 3>:
  _5 = MEM[(int (*__vtbl_ptr_type) () * *)&x];

that is only later, at ccp2 times turned int
int main() ()
{
  struct Foo x;
  long int _3;

  <bb 2>:
  _3 = (long int) Bar;
  Foo::Bar (&x);
  x ={v} {CLOBBER};
  return 0;

}

ccp1 s not able to do this because it still sees:
int main() ()
{
  struct
  {
    void Foo::<T392> (const struct Foo *) * __pfn;
    long int __delta;
  } y;
  struct Foo x;
  void Foo::<T392> (const struct Foo *) * iftmp.0;
  void Foo::<T392> (const struct Foo *) * _5;
  long int _6;
  long int _7;
  long int _9;
  sizetype _10;
  struct Foo * _11;
  int (*__vtbl_ptr_type) () * _12;
  void Foo::<T392> (const struct Foo *) * _13;
  long int _14;
  long int _15;
  sizetype _16;
  int (*__vtbl_ptr_type) () * _17;
  long int _19;
  sizetype _20;
  const struct Foo * _21;

  <bb 2>:
  y.__pfn = Bar;
  y.__delta = 0;
  _5 = y.__pfn;
  _6 = (long int) _5;
  _7 = _6 & 1;
  if (_7 == 0)
    goto <bb 3>;
  else
    goto <bb 4>;

This is fixed by SRA to:
  <bb 2>:
  y$__pfn_4 = Bar;
  y$__delta_3 = 0;
  _5 = y$__pfn_4;
  _6 = (long int) _5;
  _7 = _6 & 1;
  if (_7 == 0)
    goto <bb 3>;
  else
    goto <bb 4>;

  <bb 3>:
  iftmp.0_8 = y$__pfn_4;

Richi, is there chance for subsequent FRE to catch this?

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|jamborm at gcc dot gnu.org  |rguenth at gcc dot gnu.org

--- Comment #32 from Richard Biener <rguenth at gcc dot gnu.org> 2013-01-16 14:49:31 UTC ---
Created attachment 29181
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=29181
patch

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

--- Comment #33 from Richard Biener <rguenth at gcc dot gnu.org> 2013-03-18 08:53:51 UTC ---
Author: rguenth
Date: Mon Mar 18 08:53:42 2013
New Revision: 196771

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=196771
Log:
2013-03-18  Richard Biener  <rguenther@suse.de>

    PR tree-optimization/3713
    * tree-ssa-sccvn.c (visit_copy): Simplify.  Always propagate
    has_constants and expr.
    (stmt_has_constants): Properly valueize SSA names when deciding
    whether the stmt has constants.

    * g++.dg/ipa/devirt-12.C: New testcase.

Added:
    trunk/gcc/testsuite/g++.dg/ipa/devirt-12.C
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/tree-ssa-sccvn.c

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
      Known to work|                            |4.9.0
         Resolution|                            |FIXED

--- Comment #34 from Richard Biener <rguenth at gcc dot gnu.org> 2013-03-18 08:55:09 UTC ---
Fixed for 4.9.

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[jamborm at gcc dot gnu dot org]

------- Comment #17 from jamborm at gcc dot gnu dot org  2009-08-06 12:22 -------
Note to self: PR 40874 is related.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[jamborm at gcc dot gnu dot org]

------- Comment #16 from jamborm at gcc dot gnu dot org  2008-04-16 13:27 -------
I'm now working on inlining of indirect calls (PR 9079) and intend to allow for
inlining of calls through member pointers too.


-- 

jamborm at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |jamborm at gcc dot gnu dot
                   |dot org                     |org
             Status|NEW                         |ASSIGNED
   Last reconfirmed|2006-03-11 06:23:54         |2008-04-16 13:27:21
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[pinskia at gcc dot gnu dot org]

------- Comment #15 from pinskia at gcc dot gnu dot org  2007-09-24 19:45 -------
The testcase in comment #12 was fixed via
http://gcc.gnu.org/ml/gcc-cvs/2007-09/msg00696.html .


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[guillaume dot melquiond at ens-lyon dot fr]

------- Comment #14 from guillaume dot melquiond at ens-lyon dot fr  2007-09-02 11:56 -------
Created an attachment (id=14150)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14150&action=view)
Implements folding of "(&function & 1)"

I encountered the same issue with some heavy template code: GCC was generating
some really awful code, while it should have been optimized to almost anything.
It basically comes down to this kind of C++ code:

  struct A { void a(); };
  template < void (A::*f)() >
  void g(A *t) { (t->*f)(); }
  void h(A *t) { g<&A::a>(t); }

After instantiation, it is somehow transformed to:

  struct A { void a(); };
  void h(A *t) { (t->*(&A::a))(); }

And GCC is unable to optimize it. So I played a bit with the attached patch,
which does fix the issue for my template code.

Note that GCC does inline the member function for my code (just add an empty
body to A::a to test it), but it doesn't inline it in Andrew Pinski's last
testcase. I have no idea why; perhaps because the address goes through a
variable. But at least the generated code no longer contain any crap and it
directly calls the correct function, even if it doesn't inline it.

(As for why the C++ front-end generates this code in the first place, it is
because the lower bit of the pointer indicates whether an extra indirection is
needed when calling the function through the pointer to member. The C++
front-end actually assumes that functions have an alignment of 2 at least, so
that this trick can work.)


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[pcarlini at suse dot de]

------- Additional Comments From pcarlini at suse dot de  2005-07-04 23:42 -------
... alternately, one we really care about, performance-wise, could be distilled
from libstdc++-v3/testsuite/performance/27_io/fmtflags_manipulators.cc...

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-07-04 21:54 -------
Actually the testcase above is not really valid.
The below testcase is more correct as A::p cannot be inlined as it has its address taken (stupid bug in 
the C++ front-end really).
But here is a better testcase:
    struct A
    {
            void foo()
            {
            }
    };

    int main()
    {
     void (A::* const p)() = & A::foo;
            A a;
            (a.*p)();
    }


---- cut ---
  D.1749 = (int) foo;
  if ((D.1749 & 1) != 0) goto <L0>; else goto <L5>;

hmm, we need to know the alignment of a function, which is weird.  Why is the C++ front-end 
producing code like this, oh well.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[pinskia at gcc dot gnu dot org]

-- 
Bug 3713 depends on bug 15838, which changed state.

Bug 15838 Summary: "Inline" value of static struct
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15838

           What    |Old Value                   |New Value
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[steven at gcc dot gnu dot org]

------- Additional Comments From steven at gcc dot gnu dot org  2005-01-23 15:14 -------
*** Bug 9079 has been marked as a duplicate of this bug. ***

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-07-03 19:40 -------
*** Bug 16352 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |smelkov at mph1 dot phys dot
                   |                            |spbu dot ru


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713

[Bug tree-optimization/3713] Pointers to functions or member functions are not folded or inlined

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-06-06 01:26 -------
I filed PR15838 to track part of this bug which needs to be done to help fix this.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
  BugsThisDependsOn|15838                       |
          Component|rtl-optimization            |tree-optimization


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3713