Re: mutex_traits<>

From: bill_kempf (williamkempf_at_[hidden])
Date: 2002-01-31 16:06:41

--- In boost_at_y..., "Peter Dimov" <pdimov_at_m...> wrote:
> From: "bill_kempf" <williamkempf_at_h...>
> > I think you misunderstand me here.  A lock_ptr<> is not a thread 
safe
> > shared_ptr<>, it's a completely new smart pointer that uses a 
proxy
> > class to lock a mutex before accessing the object through 
operator->
> > and unlock the mutex afterwards.  It's an idea presented by Bjarne
> > Stroustrup, and it's very difficult to implement using only a 
Scoped
> > Lock.
> 
> A scoped lock works fine:

[snipped code]

It only works because you've removed the scoped nature by allocating 
the lock on the heap.  There are numerous reasons why people dislike 
having to use dynamic allocation in order to implement things like 
this, and in general I think they are correct.  In the early days of 
the mutex design this was brought up with the concern for 
handling "overlapping locks" and I argued for dynamic allocation 
then.  I was subsequently convinced that this wasn't good enough and 
added the lock/ulock methods to the Scoped Lock to address this.  
However, that only helps with "overlapping locks" and doesn't really 
help in cases such as lock_ptr<>.

Bill Kempf