Re: [Boost-users] Zero-malloc memory pool for Boost Intrusive smart pointers

Subject: Re: [Boost-users] Zero-malloc memory pool for Boost Intrusive smart pointers
From: Gavin Lambert (boost_at_[hidden])
Date: 2019-03-22 04:44:27

On 22/03/2019 13:09, Francesco wrote:
> I have developed for my needs a C++11 memory pool that can provides C++ 
> object instances in the form of boost::intrusive_ptr<>Â that 
> automatically return to the pool once their refcount goes to zero.
> 
> I found the need for such memory pool to have a zero-malloc memory pool 
> algorithm for some performance-oriented applications that were relying 
> on std::shared_ptr<>.
> 
> Here's the link in case anybody is interested:
> https://github.com/f18m/boost-intrusive-pool
> with some (initial) benchmark as well.

Did you look at std::make_shared<T>?  It avoids the double allocation 
for the control block that you're complaining about.  (Although you 
couldn't use it to allocate an arena block in this case -- while it does 
support array allocation, there is then only one control block for the 
whole array.)

Also the use of default construction and explicit recycle methods (and 
requiring boost_intrusive_pool_item as a base class) is ugly.

You should be able to instead keep your pool_item type internal (not 
exposed to the consumer) and use aligned_storage to hold uninitialised 
storage for the actual item data, calling T's constructor and destructor 
as needed to allocate and recycle the slot.  (Or just wrap it in 
std/boost::optional, although that can introduce some nesting ambiguities.)