Re: [boost] Re: shared_ptr/weak_ptr and thread-safety

From: Peter Dimov (pdimov_at_[hidden])
Date: 2003-05-23 07:55:06

Alexander Terekhov wrote:
> Michael Ost wrote:
> [...]
>> I wouldn't mind another BOOST_xxx precompiler setting to determine
>> thread usage in shared_ptr, ...
>
> Rather: http://google.com/groups?q=try+to+convince+Dimov+smart+policy

Or for those that don't feel like following Alexander's links:

<forward inline ;-)>

"René Dalby Carlsen" wrote:
[...]
> The above sample does not work - what should I do instead?

Try to convince Dimov that policy-based designs/interfaces aren't that bad;
and that, in general, the stuff managed by smart pointers isn't necessarily
process-wide/thread-shared stuff "by default"... like global heap, current
working directory, etc.

regards,
alexander.

--
"Yes, every user of shared_ptr that has BOOST_HAS_THREADS defined is
burdened
 with the overhead of having a shared_ptr that works. :-) shared_ptr follows
 the standard library practice here; every user is burdened with a
 synchronized heap allocator, too."
                     -- 
http://aspn.activestate.com/ASPN/Mail/Message/1263557