Boost logo

Boost :

From: Greg Colvin (gcolvin_at_[hidden])
Date: 1999-12-28 13:57:22

I recall a convincing presentation to the committee
that you can't trust library-based threading because
optimizers might reorder operations around the library
calls. That might be what Bjarne was referring to.

----- Original Message -----
From: Beman Dawes <beman_at_[hidden]>
To: <boost_at_[hidden]>; <boost_at_[hidden]>
Sent: Tuesday, December 28, 1999 8:05 AM
Subject: [boost] Re: Smart Pointers - (many) thanks and (some)questions

At 08:32 AM 12/28/99 -0500, scleary_at_[hidden] wrote:

>Should we look at doing a portable multi-threading library? I have
>beginnings of one, but it doesn't allow timed waits or waiting for
>objects -- would these be considered essential or optional?

Yes, a portable multi-threading library would be a great addition to
boost, and the C++ programming community as a whole.

You might want to approach it cautiously, however.

First, it may be more difficult that it appears. I have heard Bjarne
say several times he doesn't think it can be done properly without
language support. I don't know where he feels the problem lies, but
you might want to query him.

Second, others (POSIX, etc.) have done a great deal of work in this
area, so unless you are knowledgeable about such efforts you are
likely to waste a lot of effort reinventing the wheel.

Ideally it would be nice for someone with a strong background in
multi-treading issues were part of the team. I expect the
construction of killer test cases to particularly be a function of
hard-won experience.

If Steve and any other boost people feel up to giving it a try;
great! Go for it!


The only fruitcake at is our manager.
He's giving away a $20 coupon, plus our everyday Free Shipping.
Take advantage of the savings and selection now.

-- Check out your group's private Chat room

Boost list run by bdawes at, gregod at, cpdaniel at, john at