<!doctype html public "-//W3C//DTD W3 HTML//EN">
<html><head><style type="text/css"><!--
blockquote, dl, ul, ol, li { margin-top: 0 ; margin-bottom: 0 }
--></style><title>Re: [Python-Dev] Removing the GIL (Me, not
you!)</title></head><body>
<div>At 1:51 AM -0500 9/14/07, Justin Tulloss wrote:</div>
<blockquote type="cite" cite>On 9/14/07,<b> Adam Olsen</b> <<a
href="mailto:rhamph@gmail.com">rhamph@gmail.com</a>> wrote:<br>
<blockquote>> Could be worth a try. A first step might be to just
implement<br>
> the atomic refcounting, and run that single-threaded to see<br>
> if it has terribly bad effects on performance.<br>
<br>
I've done this experiment. It was about 12% on my
box. Later, once I<br>
had everything else setup so I could run two threads simultaneously,
I<br>
found much worse costs. All those literals become shared
objects that<br>
create contention.<br>
</blockquote>
</blockquote>
<blockquote type="cite" cite><br></blockquote>
<blockquote type="cite" cite>It's hard to argue with cold hard facts
when all we have is raw speculation. What do you think of a model
where there is a global "thread count" that keeps track of
how many threads reference an object? Then there are thread-specific
reference counters for each object. When a thread's refcount goes to
0, it decrefs the object's thread count. If you did this right,
hopefully there would only be cache updates when you update the
thread count, which will only be when a thread first references an
object and when it last references an object.</blockquote>
<div><br></div>
<div>It's likely that cache line contention is the issue, so don't
glom all the different threads' refcount for an object into one
vector. Keep each thread's refcounts in a per-thread vector of
objects, so only that thread will cache that vector, or make
refcounts so large that each will be in its own cache line (usu. 64
bytes, not too horrible for testing purposes). I don't know all
what would be required for separate vectors of refcounts, but each
object could contain its index into the vectors, which would all be
the same size (Go Virtual Memory!).</div>
<div><br></div>
<div><br></div>
<blockquote type="cite" cite>I mentioned this idea earlier and it's
growing on me. Since you've actually messed around with the code, do
you think this would alleviate some of the contention issues?<br>
</blockquote>
<blockquote type="cite" cite>Justin</blockquote>
<div><br></div>
<div>Your idea can be combined with the maxint/2 initial refcount for
non-disposable objects, which should about eliminate thread-count
updates for them.</div>
<div>-- <br>
_________________________________________<span
></span>___________________________<br>
TonyN.:' <span
></span
> <span
></span>
<mailto:tonynelson@georgeanelson.com><br>
' <span
></span
> <span
></span>
<http://www.georgeanelson.com/></div>
</body>
</html>