[Bug libstdc++/58729] New: tr2::dynamic_bitset::resize fails

[Bug libstdc++/58729] New: tr2::dynamic_bitset::resize fails

[kwbent at uab dot edu]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

            Bug ID: 58729
           Summary: tr2::dynamic_bitset::resize fails
           Product: gcc
           Version: 4.8.1
            Status: UNCONFIRMED
          Severity: minor
          Priority: P3
         Component: libstdc++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: kwbent at uab dot edu

Created attachment 31004
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=31004&action=edit
output from compilation

using g++ (Debian 4.8.1-10) 4.8.1
Debian testing x64

I've been investigating the dynamic_bitset part of the tr2 specification.  I've
used boost's dynamic bitset, so I was basing the functionality from that. 
Either I'm using the resize function wrongly, or it's failing to set the bits
as indicated.

When using the resize(number of bits, value), the resize seems to work fine,
but the value is ignored.  I started to dig around; the resize function is
defined as this in the header <tr2/dynamic_bitset>

      void
      resize(size_type __nbits, bool __value = false)
      {
    this->_M_resize(__nbits, __value);
    this->_M_Nb = __nbits;
    this->_M_do_sanitize();
      }

Is the value being defaulted to false, regardless of the input?  The actual
main() is at the end of the main.ii file.  Another oddity is that this will not
compile from the command line, but will from inside netbeans where I code, even
with Wall and Wextra specified.

If more info is needed, I'll oblige.

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[kwbent at uab dot edu]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

--- Comment #1 from Kyle Bentley <kwbent at uab dot edu> ---
Created attachment 31005
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=31005&action=edit
output from the compile(as seen from terminal)

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[kwbent at uab dot edu]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

--- Comment #2 from Kyle Bentley <kwbent at uab dot edu> ---
Created attachment 31006
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=31006&action=edit
Source code

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[kwbent at uab dot edu]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

--- Comment #3 from Kyle Bentley <kwbent at uab dot edu> ---
Created attachment 31007
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=31007&action=edit
Code output

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2013-10-14
     Ever confirmed|0                           |1

--- Comment #4 from Paolo Carlini <paolo.carlini at oracle dot com> ---
First blush the problem is in _M_resize: it ignores the __value passed as
argument. That can't be right. Ed can you have a look?

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[3dw4rd at verizon dot net]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

--- Comment #5 from Ed Smith-Rowland <3dw4rd at verizon dot net> ---
This is wrong.  Testing a patch...

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[3dw4rd at verizon dot net]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

--- Comment #6 from Ed Smith-Rowland <3dw4rd at verizon dot net> ---
Created attachment 31017
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=31017&action=edit
Patch dynamic_bitset...

Here is the patch.  I did a little extra work to move large functions offline
and to fix other things that were broken.  This class was dumped in before it
was ready really.

tested on x86_64-linux.

I tried retesting this morning but it looks like bootstrap is broken.

See what you think.

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[redi at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

--- Comment #7 from Jonathan Wakely <redi at gcc dot gnu.org> ---
Comment on attachment 31017
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=31017
Patch dynamic_bitset...

The new dynamic_bitset.tcc file should have a different doxygen block from the
main header:

+/** @file tr2/dynamic_bitset
+ *  This is a TR2 C++ Library header.
+ */

I suggest:

/** @file tr2/dynamic_bitset.tcc
 *  This is an internal header file, included by other library headers.
 *  Do not attempt to use it directly. @headername{tr2/dynamic_bitset}

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

--- Comment #8 from Paolo Carlini <paolo.carlini at oracle dot com> ---
By the way, I just fixed the bootstrap. When you are ready, please send the
patch to the mailing list, thanks!

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[redi at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

--- Comment #9 from Jonathan Wakely <redi at gcc dot gnu.org> ---
Also, your doxygen markup is not valid, this does not start a group:

+  //@{
+
+  /**
+   *  @brief Global I/O operators for bitsets.
+   *

That just adds the @brief to the next function, I think you want:

  /**
   *  @defgroup Global I/O operators for bitsets.
   *  @{
   */

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[3dw4rd at verizon dot net]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="UTF-8", Size: 3071 bytes --]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

--- Comment #10 from Ed Smith-Rowland <3dw4rd at verizon dot net> ---
The pt.c fix worked but...

Still getting:
In file included from ../../gcc/gcc/lto/lto.c:3301:0:
./gt-lto-lto.h:153:6: error: ‘type_hash_cache’ was not declared in this scope
./gt-lto-lto.h:155:13: error: ‘type_hash_cache’ was not declared in this scope
./gt-lto-lto.h:161:6: error: ‘gimple_types’ was not declared in this scope
./gt-lto-lto.h:163:13: error: ‘gimple_types’ was not declared in this scope
./gt-lto-lto.h:173:6: error: ‘type_hash_cache’ was not declared in this scope
./gt-lto-lto.h:175:13: error: ‘type_hash_cache’ was not declared in this scope
./gt-lto-lto.h:180:6: error: ‘gimple_types’ was not declared in this scope
./gt-lto-lto.h:182:13: error: ‘gimple_types’ was not declared in this scope

I have to run.  I'll try again tonight.
>From gcc-bugs-return-431934-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Wed Oct 16 12:58:01 2013
Return-Path: <gcc-bugs-return-431934-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 9083 invoked by alias); 16 Oct 2013 12:58:00 -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 9005 invoked by uid 48); 16 Oct 2013 12:57:58 -0000
From: "jakub at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug sanitizer/56393] SIGSEGV when -fsanitize­dress and dynamic lib with global objects
Date: Wed, 16 Oct 2013 12:58:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: sanitizer
X-Bugzilla-Version: 4.8.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: jakub at gcc dot gnu.org
X-Bugzilla-Status: ASSIGNED
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-56393-4-CCo8CheYur@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-56393-4@http.gcc.gnu.org/bugzilla/>
References: <bug-56393-4@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: 2013-10/txt/msg01078.txt.bz2
Content-length: 359

http://gcc.gnu.org/bugzilla/show_bug.cgi?idV393

--- Comment #36 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
And the advantage of that patch would be?  Previously ASAN would often work
even when binary wasn't linked with -fsanitize­dress, though sometimes it
wouldn't, while with the patch it would never work.  That I don't see as an
improvement.

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

--- Comment #11 from Paolo Carlini <paolo.carlini at oracle dot com> ---
I don't see that. Boot & test just completed successfully for me.

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

--- Comment #12 from Paolo Carlini <paolo.carlini at oracle dot com> ---
In my x86_64 -m64 build, gt-lto-lto.h has only 149 lines. Looks like your tree
is in an inconsistent state.

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[emsr at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

--- Comment #13 from emsr at gcc dot gnu.org ---
Author: emsr
Date: Sat Oct 19 01:31:19 2013
New Revision: 203841

URL: http://gcc.gnu.org/viewcvs?rev=203841&root=gcc&view=rev
Log:
2013-10-18  Edward Smith-Rowland  <3dw4rd@verizon.net>

    PR libstdc++/58729
    * include/tr2/dynamic_bitset (_M_resize, resize): Use input value
    to set bits; (_M_do_left_shift, _M_do_right_shift, _M_do_to_ulong,
    _M_do_to_ullong, _M_do_find_first, _M_do_find_next, _M_copy_from_ptr,
    operator>>): Move long methods outline to...
    * include/tr2/dynamic_bitset.tcc: New.
    * include/Makefile.am: Add dynamic_bitset.tcc.
    * include/Makefile.in: Add dynamic_bitset.tcc.
    * testsuite/tr2/dynamic_bitset/pr58729.cc: New.


Added:
    trunk/libstdc++-v3/include/tr2/dynamic_bitset.tcc
    trunk/libstdc++-v3/testsuite/tr2/dynamic_bitset/
    trunk/libstdc++-v3/testsuite/tr2/dynamic_bitset/pr58729.cc
Modified:
    trunk/libstdc++-v3/ChangeLog
    trunk/libstdc++-v3/include/Makefile.am
    trunk/libstdc++-v3/include/Makefile.in
    trunk/libstdc++-v3/include/tr2/dynamic_bitset

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[3dw4rd at verizon dot net]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

--- Comment #14 from Ed Smith-Rowland <3dw4rd at verizon dot net> ---
See http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58804 for work that will get
this to work for -m32 also.

I was using builtins for long rather than long long so the last patch failed
with -m32.  Sigh...

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[emsr at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

--- Comment #15 from emsr at gcc dot gnu.org ---
Author: emsr
Date: Mon Oct 21 13:52:39 2013
New Revision: 203893

URL: http://gcc.gnu.org/viewcvs?rev=203893&root=gcc&view=rev
Log:
2013-10-20  Edward Smith-Rowland  <3dw4rd@verizon.net>

    PR libstdc++/58804
    PR libstdc++/58729
    * include/tr2/dynamic_bitset
    (__dynamic_bitset_base<_WordT, _Alloc>::_M_are_all_aux,
    __dynamic_bitset_base<_WordT, _Alloc>::_M_do_count):
    Use __builtin_popcountll() instead of __builtin_popcountl().
    * include/tr2/dynamic_bitset.tcc
    (__dynamic_bitset_base<_WordT, _Alloc>::_M_do_find_first,
    __dynamic_bitset_base<_WordT, _Alloc>::_M_do_find_next):
    Use __builtin_ctzll() instead of __builtin_ctzl().


Modified:
    trunk/libstdc++-v3/ChangeLog
    trunk/libstdc++-v3/include/tr2/dynamic_bitset
    trunk/libstdc++-v3/include/tr2/dynamic_bitset.tcc

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[paolo.carlini at oracle dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

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

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

--- Comment #16 from Paolo Carlini <paolo.carlini at oracle dot com> ---
Fixed.

[Bug libstdc++/58729] tr2::dynamic_bitset::resize fails

[redi at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58729

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.9.0