Re: Lock free containers.

[Phillip Jordan <phillip dot m dot jordan at gmail dot com>]

Dhruv Matani wrote:
>  Thanks for the detailed plan. However, I did notice that you have
> missed out hash_[map|set|multimap|multiset] from the stuff that you
> intend working on for making lock free(or at least reentrant for
> certain key operations). I admit that I do have some personal vested
> interest in the hash family of containers becoming lock free(if
> possible) since I use them quite extensively. So, if possible, it
> would be great to have that working too.

As of TR1, the hash-table-based associative containers are standard, and
have the names unordered_set, unordered_map, unordered_multimap, and
unordered_multiset, in the std::tr1 namespace. As I said,

> On 5/29/06, Phillip Jordan <phillip.m.jordan@gmail.com> wrote:
>> This will probably keep me busy for a while anyway, but hopefully I can
>> work on some of the other containers, too, std::deque and std::vector
>> would be very useful, and I think std::tr1::unordered_* can benefit as
>> well.

I do plan to work on these, they just won't be the first ones I'll be
looking at, though they shouldn't be too difficult to adapt, apart from
the resizing issue...

I hope that clears things up.

~phil