Re: [boost] portable precise garbage collector for C++

From: Achilleas Margaritis (axilmar_at_[hidden])
Date: 2007-09-23 16:45:16

O/H Larry Evans έγραψε:
> On 09/23/07 13:31, Achilleas Margaritis wrote:
>> O/H Larry Evans έγραψε:
> [snip]
>>> The main_stl.zip in boost-vault/memory uses Sebastian's suggestion and
>>> it seems to work.  It runs a little std_list test, then a GCBench
>>> with std:pair, then GCBench with std_list<., gc_allocator>.
>>>
>>> As (I think) you're pointing out above, the std_list<T,gc_allocator<T> >
>>> would define gc_allocator<T>::pointer as gc_ptr<T>, which is
>>> never used in std_list at all because the std_list stores the
>>> complete T (not T*) in std_list<...>::node.  I can see where this
>>> would be confusing and is a shortcoming of the allocator model.
>>> It just happens to work in this case.
>> I am not really following you. Does your STL list class use the class 
>> gc_allocator<T> or not? and if it does, does it use it fully for all its 
>> operations or for some of them?
> 
> It can, just like std::list can.  IOW, it can be the 2nd template
> parameter.  The list_pair template (derived from std_list) uses
> gc_allocator as the 2nd template parameter of its superclass,
> std_list<Node,gc_allocator<Node> >.
> 
> It doesn't use it for any operations because
> it's part of the nested  std::list<.,.>::node used to implement the
> list.  Only push_back was coded since that's the only one used.
> I guess the std_list::empty is not standard; however, that's
> not really important since the only purpose was to demonstrate
> how, possibly a future std::list, could be coded to allow
> your gc_allocator to be used.  Now that I think about it,
> maybe only the std_list::node::_next needs to be a gc_ptr<.>.
> That would save some time.
> 

Ah, ok, now I get it :-). You are coding your own list.

All you have to do is make every pointer gc_ptr<T> and allocate every 
object using the collector.

The node object must be allocated using gc_new.