[boost] OOPS! and where's _page::operator new called (was Re: portable precise garbage collector for C++

From: Larry Evans (cppljevans_at_[hidden])
Date: 2007-09-19 15:37:08

On 09/19/07 14:11, Larry Evans wrote:
[snip]
>>> main_test only contains:
>>>
>>> #include "cppgc.hpp"
>>> #include <iostream>
>>> int main(void)
>>> {
>>>      void*p=dlmalloc(20);
>>>      void*a_page=_page::operator new(sizeof(_page));
>>>      std::cout<<"running "<<__FILE__<<"\n";
>>>      std::cout<<"p="<<p<<"\n";
>>>      std::cout<<"a_page="<<a_page<<"\n";
>>>      return 0;
>>> }
>> I removed dlmalloc...I will try to compile it under Linux as soon as 
>> possible.
> 
> I'm mystified by what's happening because when I changed main to:
> 
>        _page*a_page=new _page;
> 
> I got the segfault again even when the _page::operator new had:
> 
>      void*p=
>      #if 0
>        _CPPGC_MALLOC(gc::_mem_space, size);
>      #else
>        0;
>      #endif
> 
> ???
OOPS.  Maybe because a _page cannot be constructed at location 0 :)

Anyway, I tried running with:

   void*p=::operator new(size);

and it worked, then tried with src/main.cpp but it never terminated
until I manually aborted.

Hope you can get the unix working soon.

Good luck!