[Bug c++/60792] New: bogus buffer overflow warning and abort on static flexible array member in a child object

[Bug c++/60792] New: bogus buffer overflow warning and abort on static flexible array member in a child object

[abalint21 at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60792

            Bug ID: 60792
           Summary: bogus buffer overflow warning and abort on static
                    flexible array member in a child object
           Product: gcc
           Version: 4.8.2
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: abalint21 at gmail dot com

g++ emits a bogus warning on the program below which then aborts at runtime.
The strange thing is that if I get the reference of the child object and then
get the address of the str field then everything is OK. It seems that g++
cannot handle the inner child object's str if it is accessed via
parent->child.str but it is ok when a reference is taken from the child and
then accessed via child.str.


$ cat main.cpp && g++ -D_FORTIFY_SOURCE=2 -O2 main.cpp && ./a.out 
#include <cstdlib>
#include <cstring>
#include <iostream>

struct Parent
{

    struct Child
    {
        int a;
        char b;
        char str[0]; ///< ASCIIZ
    } child;
};

//#define DONT_CRASH

int main(int argc, char** argv)
{
    char* buffer = new char[32768];

    Parent* parent = (Parent*) buffer;

    parent->child.a = 1;
    parent->child.b = 'a';

#ifdef DONT_CRASH
    Parent::Child& child = parent->child;
    char* childStr = child.str;
#else
    char* childStr = parent->child.str;
#endif

    std::cout << __USE_FORTIFY_LEVEL << std::endl;
    std::cout << __bos(childStr) << std::endl;

    size_t strLen = 4;
    std::strncpy(childStr, "test", strLen);
    if (childStr[strLen] not_eq '\0')
    {
        childStr[strLen] = '\0';
    }

    return 0;
}

In file included from /usr/include/string.h:640:0,
                 from /usr/include/c++/4.8.2/cstring:42,
                 from main.cpp:2:
In function ‘char* strncpy(char*, const char*, size_t)’,
    inlined from ‘int main(int, char**)’ at main.cpp:38:43:
/usr/include/bits/string3.h:120:71: warning: call to char*
__builtin___strncpy_chk(char*, const char*, long unsigned int, long unsigned
int) will always overflow destination buffer [enabled by default]
   return __builtin___strncpy_chk (__dest, __src, __len, __bos (__dest));
                                                                       ^
2
0
*** buffer overflow detected ***: ./a.out terminated
...
Aborted
>From gcc-bugs-return-448593-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Wed Apr 09 10:28:11 2014
Return-Path: <gcc-bugs-return-448593-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 7985 invoked by alias); 9 Apr 2014 10:28:11 -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 7947 invoked by uid 48); 9 Apr 2014 10:28:08 -0000
From: "gnugcc at marino dot st" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug ada/60730] 'Round of a fixed point type incorrectly truncates its operand instead of rounding it
Date: Wed, 09 Apr 2014 10:28:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: ada
X-Bugzilla-Version: 4.9.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: gnugcc at marino dot st
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: cc
Message-ID: <bug-60730-4-ktbfzUzq0U@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-60730-4@http.gcc.gnu.org/bugzilla/>
References: <bug-60730-4@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: 2014-04/txt/msg00613.txt.bz2
Content-length: 443

http://gcc.gnu.org/bugzilla/show_bug.cgi?id`730

John Marino <gnugcc at marino dot st> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |gnugcc at marino dot st

--- Comment #2 from John Marino <gnugcc at marino dot st> ---
what platform is this program being run on?
what is output of "uname -a" ?

[Bug c++/60792] bogus buffer overflow warning and abort on flexible array member in a child object

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60792

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

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

--- Comment #1 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Zero length array inside of another string is not a flexible array member, nor
anything close to that, and with -D_FORTIFY_SOURCE=2 is not even supposed to be
handled like flexible array member alternative.  Thus, I don't see why you
think this is a bug.  Simply don't do it, use -D_FORTIFY_SOURCE=1 instead, or
use e.g. memcpy instead of strncpy which is allowed even in -D_FORTIFY_SOURCE=2
mode (which is stricter than what C/C++ allows) to cross field boundaries.

[Bug c++/60792] bogus buffer overflow warning and abort on flexible array member in a child object

[abalint21 at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60792

Attila Balint <abalint21 at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|---                         |INVALID

--- Comment #2 from Attila Balint <abalint21 at gmail dot com> ---
(In reply to Jakub Jelinek from comment #1)
> Zero length array inside of another string is not a flexible array member,
> nor anything close to that, and with -D_FORTIFY_SOURCE=2 is not even
> supposed to be handled like flexible array member alternative.  Thus, I
> don't see why you think this is a bug.  Simply don't do it, use
> -D_FORTIFY_SOURCE=1 instead, or use e.g. memcpy instead of strncpy which is
> allowed even in -D_FORTIFY_SOURCE=2 mode (which is stricter than what C/C++
> allows) to cross field boundaries.

Clear. Thank you for the clarification. It is not a bug