<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
	xmlns:georss="http://www.georss.org/georss" xmlns:geo="http://www.w3.org/2003/01/geo/wgs84_pos#" xmlns:media="http://search.yahoo.com/mrss/"
		>
<channel>
	<title>Comments on: jemalloc now on the trunk</title>
	<atom:link href="http://blog.pavlov.net/2008/02/05/jemalloc-now-on-the-trunk/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.pavlov.net/2008/02/05/jemalloc-now-on-the-trunk/</link>
	<description>Ramblings from the mind of Stuart Parmenter</description>
	<lastBuildDate>Thu, 28 Jan 2010 21:51:52 +0000</lastBuildDate>
	<generator>http://wordpress.com/</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: Cory Nelson</title>
		<link>http://blog.pavlov.net/2008/02/05/jemalloc-now-on-the-trunk/#comment-2852</link>
		<dc:creator>Cory Nelson</dc:creator>
		<pubDate>Wed, 20 Feb 2008 23:14:18 +0000</pubDate>
		<guid isPermaLink="false">http://pavlovdotnet.wordpress.com/?p=458#comment-2852</guid>
		<description>After playing around with the config string I got it up to about 70% scalability - a good 10% more than Windows&#039; heaps.  The tricky thing was that despite being more scalable it performed poorly enough - contention or not - that overall it was still a big loss in throughput.

The big perf issue appears to be in arena_purge - increasing the F flag fixed that nicely.  Some scalability was lost and memory usage has increased very slightly, but overall performance is now 5-10% better than Windows&#039; heaps.

I noticed JE removed lazy free completely from FBSD&#039;s CVS, citing performance reasons.  I have to agree - I ported it over and it has been a detriment in most ways I tried to use it.</description>
		<content:encoded><![CDATA[<p>After playing around with the config string I got it up to about 70% scalability &#8211; a good 10% more than Windows&#8217; heaps.  The tricky thing was that despite being more scalable it performed poorly enough &#8211; contention or not &#8211; that overall it was still a big loss in throughput.</p>
<p>The big perf issue appears to be in arena_purge &#8211; increasing the F flag fixed that nicely.  Some scalability was lost and memory usage has increased very slightly, but overall performance is now 5-10% better than Windows&#8217; heaps.</p>
<p>I noticed JE removed lazy free completely from FBSD&#8217;s CVS, citing performance reasons.  I have to agree &#8211; I ported it over and it has been a detriment in most ways I tried to use it.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Friendly_GUI</title>
		<link>http://blog.pavlov.net/2008/02/05/jemalloc-now-on-the-trunk/#comment-2847</link>
		<dc:creator>Friendly_GUI</dc:creator>
		<pubDate>Thu, 14 Feb 2008 04:44:25 +0000</pubDate>
		<guid isPermaLink="false">http://pavlovdotnet.wordpress.com/?p=458#comment-2847</guid>
		<description>Could somebody post a recommended testing procedures so we can compare the performance difference?

A bookmark file and program to use etc...

Thanks</description>
		<content:encoded><![CDATA[<p>Could somebody post a recommended testing procedures so we can compare the performance difference?</p>
<p>A bookmark file and program to use etc&#8230;</p>
<p>Thanks</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Matt Doran</title>
		<link>http://blog.pavlov.net/2008/02/05/jemalloc-now-on-the-trunk/#comment-2845</link>
		<dc:creator>Matt Doran</dc:creator>
		<pubDate>Tue, 12 Feb 2008 11:26:37 +0000</pubDate>
		<guid isPermaLink="false">http://pavlovdotnet.wordpress.com/?p=458#comment-2845</guid>
		<description>Did you get a chance to update measurements of memory, performance and fragmentation?

I&#039;m really curious to see how much a difference these improvements made!</description>
		<content:encoded><![CDATA[<p>Did you get a chance to update measurements of memory, performance and fragmentation?</p>
<p>I&#8217;m really curious to see how much a difference these improvements made!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: pd</title>
		<link>http://blog.pavlov.net/2008/02/05/jemalloc-now-on-the-trunk/#comment-2835</link>
		<dc:creator>pd</dc:creator>
		<pubDate>Wed, 06 Feb 2008 22:52:15 +0000</pubDate>
		<guid isPermaLink="false">http://pavlovdotnet.wordpress.com/?p=458#comment-2835</guid>
		<description>mozcrt19.dll ? Got it. :)</description>
		<content:encoded><![CDATA[<p>mozcrt19.dll ? Got it. <img src='http://s.wordpress.com/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /> </p>
]]></content:encoded>
	</item>
	<item>
		<title>By: RyanVM</title>
		<link>http://blog.pavlov.net/2008/02/05/jemalloc-now-on-the-trunk/#comment-2834</link>
		<dc:creator>RyanVM</dc:creator>
		<pubDate>Wed, 06 Feb 2008 18:10:15 +0000</pubDate>
		<guid isPermaLink="false">http://pavlovdotnet.wordpress.com/?p=458#comment-2834</guid>
		<description>pd, Ferdinand - If you&#039;ve got mozcrt19.dll in your Minefield directory, you&#039;ve got a jemalloc-enabled build.

That said, Stuart, is there a bug filed for the DLL not having proper resource information set on it (version info and the like)? Also, I saw a comment on another blog (I think it was bsmedberg&#039;s) about possibly linking the CRT with xul.dll. Is that a worthwhile possibility or a waste of time?</description>
		<content:encoded><![CDATA[<p>pd, Ferdinand &#8211; If you&#8217;ve got mozcrt19.dll in your Minefield directory, you&#8217;ve got a jemalloc-enabled build.</p>
<p>That said, Stuart, is there a bug filed for the DLL not having proper resource information set on it (version info and the like)? Also, I saw a comment on another blog (I think it was bsmedberg&#8217;s) about possibly linking the CRT with xul.dll. Is that a worthwhile possibility or a waste of time?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Ferdinand</title>
		<link>http://blog.pavlov.net/2008/02/05/jemalloc-now-on-the-trunk/#comment-2833</link>
		<dc:creator>Ferdinand</dc:creator>
		<pubDate>Wed, 06 Feb 2008 13:19:18 +0000</pubDate>
		<guid isPermaLink="false">http://pavlovdotnet.wordpress.com/?p=458#comment-2833</guid>
		<description>Does this mean that the newest nightly&#039;s already use jemalloc? So everybody with 2008020504 should see better performance and less memory usage in the windows taskmanager? Or is it only used if you build it yourself?</description>
		<content:encoded><![CDATA[<p>Does this mean that the newest nightly&#8217;s already use jemalloc? So everybody with 2008020504 should see better performance and less memory usage in the windows taskmanager? Or is it only used if you build it yourself?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Mark S</title>
		<link>http://blog.pavlov.net/2008/02/05/jemalloc-now-on-the-trunk/#comment-2832</link>
		<dc:creator>Mark S</dc:creator>
		<pubDate>Wed, 06 Feb 2008 07:06:41 +0000</pubDate>
		<guid isPermaLink="false">http://pavlovdotnet.wordpress.com/?p=458#comment-2832</guid>
		<description>So let&#039;s say we decide not to use jemalloc on mac (or linux).
Would we make this decision because the perf &amp; size wins are &#039;not enough&#039;?  If there is some, wouldn&#039;t it still be worth it?  Is there some other detractor about using it that would make it not worth turning on?
Also, if it were not turned on for mac would an investigation ensue to use something else for bigger wins?  (For that matter, if it *is* turned on will we still look for an alternative that gets better wins?)</description>
		<content:encoded><![CDATA[<p>So let&#8217;s say we decide not to use jemalloc on mac (or linux).<br />
Would we make this decision because the perf &amp; size wins are &#8216;not enough&#8217;?  If there is some, wouldn&#8217;t it still be worth it?  Is there some other detractor about using it that would make it not worth turning on?<br />
Also, if it were not turned on for mac would an investigation ensue to use something else for bigger wins?  (For that matter, if it *is* turned on will we still look for an alternative that gets better wins?)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: alanjstr</title>
		<link>http://blog.pavlov.net/2008/02/05/jemalloc-now-on-the-trunk/#comment-2831</link>
		<dc:creator>alanjstr</dc:creator>
		<pubDate>Wed, 06 Feb 2008 05:15:57 +0000</pubDate>
		<guid isPermaLink="false">http://pavlovdotnet.wordpress.com/?p=458#comment-2831</guid>
		<description>Gimme gimme gimme gimme!  I don&#039;t wanna wait until beta 4 :-(.  But the only Firefox 3 I use is the portable version (currently b3).  I&#039;m excited for FF3, but I&#039;ll have to get over some of the UI changes.</description>
		<content:encoded><![CDATA[<p>Gimme gimme gimme gimme!  I don&#8217;t wanna wait until beta 4 <img src='http://s.wordpress.com/wp-includes/images/smilies/icon_sad.gif' alt=':-(' class='wp-smiley' /> .  But the only Firefox 3 I use is the portable version (currently b3).  I&#8217;m excited for FF3, but I&#8217;ll have to get over some of the UI changes.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: pd</title>
		<link>http://blog.pavlov.net/2008/02/05/jemalloc-now-on-the-trunk/#comment-2830</link>
		<dc:creator>pd</dc:creator>
		<pubDate>Wed, 06 Feb 2008 03:35:41 +0000</pubDate>
		<guid isPermaLink="false">http://pavlovdotnet.wordpress.com/?p=458#comment-2830</guid>
		<description>Stuart could you please confirm the build number just so we can be sure we&#039;re testing the jemalloc-enabled version? I just downloaded build 2008020504, is that the one?</description>
		<content:encoded><![CDATA[<p>Stuart could you please confirm the build number just so we can be sure we&#8217;re testing the jemalloc-enabled version? I just downloaded build 2008020504, is that the one?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Cory Nelson</title>
		<link>http://blog.pavlov.net/2008/02/05/jemalloc-now-on-the-trunk/#comment-2829</link>
		<dc:creator>Cory Nelson</dc:creator>
		<pubDate>Wed, 06 Feb 2008 03:06:07 +0000</pubDate>
		<guid isPermaLink="false">http://pavlovdotnet.wordpress.com/?p=458#comment-2829</guid>
		<description>Ah, I did not realize that.  I will have to play with the options a bit more then.</description>
		<content:encoded><![CDATA[<p>Ah, I did not realize that.  I will have to play with the options a bit more then.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
