Re: ada/9273: g++ 3.2 generates segmentation fault for correct code

Re: ada/9273: g++ 3.2 generates segmentation fault for correct code

[John Ahlgren]

The following reply was made to PR ada/9273; it has been noted by GNATS.

>From john.ahlgren@home.se  Mon Jan 13 00:57:20 2003
From: John Ahlgren <john.ahlgren@home.se>
To: gcc-gnats@gcc.gnu.org
Cc:  
Subject: Re: ada/9273: g++ 3.2 generates segmentation fault for correct code
Date: 11 Jan 2003 09:50:32 +0100

 Hi,
 
 I'm sorry about my bug report; it is invalid.
 I made the assumption that no object of A was copied, but that
 was the case. So, may I now bother you asking why it is copied:
 class A { ... };
 A a;
 vector<A> avec;
 // avec.reserve(2) (*)
 avec.push_back(a);
 avec.push_back(a);
 
 With (*) of course nothing is copied, but even so, can't realloc() be
 used instead?
 
 Cheers,
 John
 
 
 
 On Sat, 2003-01-11 at 09:06, gcc-gnats@gcc.gnu.org wrote:
 > Thank you very much for your problem report.
 > It has the internal identification `ada/9273'.
 > The individual assigned to look at your
 > report is: unassigned. 
 > 
 > >Category:       ada
 > >Responsible:    unassigned
 > >Synopsis:       g++ 3.2 generates segmentation fault for correct code
 > >Arrival-Date:   Sat Jan 11 00:06:00 PST 2003
 > 
 
 

ada/9273: g++ 3.2 generates segmentation fault for correct code

[john.ahlgren]

>Number:         9273
>Category:       ada
>Synopsis:       g++ 3.2 generates segmentation fault for correct code
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          wrong-code
>Submitter-Id:   net
>Arrival-Date:   Sat Jan 11 00:06:00 PST 2003
>Closed-Date:
>Last-Modified:
>Originator:     John Ahlgren
>Release:        gcc version 3.2 20020903 (Red Hat Linux 8.0 3.2-7)
>Organization:
>Environment:
x386, Redhat 8.0 with Linux kernel 2.4.18-19.8.0
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --disable-checking --host=i386-redhat-linux --with-system-zlib --enable-__cxa_atexit
Thread model: posix
>Description:
#include <iostream>
#include <string>
#include <vector>

using namespace std;

class A {
public:
  A(string s) { sptr = new string; }
  ~A(void) { cerr << "delete " << sptr << "\n"; delete sptr; cerr << "delete done\n"; }
private:
  string* sptr;
};

int main()
{
  vector<A> avec;
  A a("a");
  avec.push_back(a);
  avec.push_back(a);

  cerr << "main done\n";
}
>How-To-Repeat:
Compile with g++ -Wall file -o test (no warnings are output)
Then just run it, it will segfault trying to delete the pointer A::sptr

Is this really a bug?
>Fix:
Unfortunately, I have no idea :-(
>Release-Note:
>Audit-Trail:
>Unformatted: