Re: better atomics - v0.6 - Mailing list pgsql-hackers

From Andres Freund
Subject Re: better atomics - v0.6
Date
Msg-id 20140926010840.GJ16581@awork2.anarazel.de
Whole thread Raw
In response to Re: better atomics - v0.6  (Robert Haas <robertmhaas@gmail.com>)
List pgsql-hackers
On 2014-09-25 21:02:28 -0400, Robert Haas wrote:
> I feel like this could really use a developer README: what primitives
> do we have, which ones are likely to be efficient or inefficient on
> which platforms, how do atomics interact with barriers, etc.

atomics.h has most of that. Including documenting which barrier
semantics the individual operations have. One change I can see as being
a good idea is to move/transform the definition of acquire/release
barriers from s_lock.h to README.barrier.

It doesn't document which operations are efficient on which platform,
but I wouldn't know what to say about that topic? The current set of
operations should be fast on pretty much all platforms.

What information would you like to see? An example of how to use
atomics?

Greetings,

Andres Freund

-- Andres Freund                       http://www.2ndQuadrant.com/PostgreSQL Development, 24x7 Support, Training &
Services



pgsql-hackers by date:

Previous
From: Robert Haas
Date:
Subject: Re: better atomics - v0.6
Next
From: Bruce Momjian
Date:
Subject: Re: KNN-GiST with recheck