<?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/"
		>
<channel>
	<title>Comments for Saturn Valley</title>
	<atom:link href="http://blog.mozilla.com/adw/comments/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.mozilla.com/adw</link>
	<description></description>
	<lastBuildDate>Mon, 07 Sep 2009 19:45:58 -0700</lastBuildDate>
	<generator>http://wordpress.org/?v=2.8.6</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>Comment on Cold Ts by Zack</title>
		<link>http://blog.mozilla.com/adw/2009/09/04/cold-ts/#comment-129</link>
		<dc:creator>Zack</dc:creator>
		<pubDate>Mon, 07 Sep 2009 19:45:58 +0000</pubDate>
		<guid isPermaLink="false">http://blog.mozilla.com/adw/?p=72#comment-129</guid>
		<description>Like so:

#include 
#include 

int
main(void)
{
  int fd = open(&quot;/proc/sys/vm/drop_caches&quot;, O_WRONLY);
  if (fd == -1) return 1;
  write(fd, &quot;3\n&quot;, 2);
  close(fd);
  return 0;
}


If you go this way I strongly suggest using dietlibc (http://www.fefe.de/dietlibc/) as glibc&#039;s support for static linking is incredibly half-assed these days.  dietlibc will produce a binary that you can easily disassemble and inspect for safety.</description>
		<content:encoded><![CDATA[<p>Like so:</p>
<p>#include<br />
#include </p>
<p>int<br />
main(void)<br />
{<br />
  int fd = open(&#8221;/proc/sys/vm/drop_caches&#8221;, O_WRONLY);<br />
  if (fd == -1) return 1;<br />
  write(fd, &#8220;3\n&#8221;, 2);<br />
  close(fd);<br />
  return 0;<br />
}</p>
<p>If you go this way I strongly suggest using dietlibc (<a href="http://www.fefe.de/dietlibc/" rel="nofollow">http://www.fefe.de/dietlibc/</a>) as glibc&#8217;s support for static linking is incredibly half-assed these days.  dietlibc will produce a binary that you can easily disassemble and inspect for safety.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Cold Ts by Zack</title>
		<link>http://blog.mozilla.com/adw/2009/09/04/cold-ts/#comment-128</link>
		<dc:creator>Zack</dc:creator>
		<pubDate>Mon, 07 Sep 2009 19:26:51 +0000</pubDate>
		<guid isPermaLink="false">http://blog.mozilla.com/adw/?p=72#comment-128</guid>
		<description>Or you could write a short C program that does the equivalent of &quot;echo 3 &gt; /proc/sys/vm/drop_caches&quot; and install it setuid to root, executable only by a group that includes the talos user (but, ideally, no one else).</description>
		<content:encoded><![CDATA[<p>Or you could write a short C program that does the equivalent of &#8220;echo 3 &gt; /proc/sys/vm/drop_caches&#8221; and install it setuid to root, executable only by a group that includes the talos user (but, ideally, no one else).</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Cold Ts by bws42</title>
		<link>http://blog.mozilla.com/adw/2009/09/04/cold-ts/#comment-127</link>
		<dc:creator>bws42</dc:creator>
		<pubDate>Sun, 06 Sep 2009 15:20:58 +0000</pubDate>
		<guid isPermaLink="false">http://blog.mozilla.com/adw/?p=72#comment-127</guid>
		<description>I&#039;ll reply for anon. If you look at the man page for the sudoers config file it will show you the line you can add to allow the talos account to execute a single command without a password prompt. The line would look something like this:

talos talos_machine = NOPASSWD: /path/to/script/to/clear/cache

where the script would be owned by root to prevent modification. I used a similar setup when I need a web interface to reload firewall rules after editing.</description>
		<content:encoded><![CDATA[<p>I&#8217;ll reply for anon. If you look at the man page for the sudoers config file it will show you the line you can add to allow the talos account to execute a single command without a password prompt. The line would look something like this:</p>
<p>talos talos_machine = NOPASSWD: /path/to/script/to/clear/cache</p>
<p>where the script would be owned by root to prevent modification. I used a similar setup when I need a web interface to reload firewall rules after editing.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Cold Ts by adw</title>
		<link>http://blog.mozilla.com/adw/2009/09/04/cold-ts/#comment-126</link>
		<dc:creator>adw</dc:creator>
		<pubDate>Sat, 05 Sep 2009 23:46:50 +0000</pubDate>
		<guid isPermaLink="false">http://blog.mozilla.com/adw/?p=72#comment-126</guid>
		<description>Thanks for the comments everyone.

@Anon #1: Is there a way to suppress the password prompt without compromising security?  Could you point me to some documentation?

@Robert: Sure, thanks for the idea.  I&#039;ll get to it next week.

@ant: A kind soul has already written a &lt;a href=&quot;http://chadaustin.me/2009/04/flushing-disk-cache/&quot; rel=&quot;nofollow&quot;&gt;program&lt;/a&gt; that does just that.  (It&#039;s listed on the notes page I linked above.)  It does work, but problem is, it grinds your system to a halt for ten minutes as it does its business.  The more RAM, the longer it takes.  Ts is the average of 20 runs, so that&#039;s three hours and twenty minutes just to complete one Ts.  Maybe we&#039;ll have to use it, though.

@Anon #2: Sure, I meant that we would have caught it as we landed it.</description>
		<content:encoded><![CDATA[<p>Thanks for the comments everyone.</p>
<p>@Anon #1: Is there a way to suppress the password prompt without compromising security?  Could you point me to some documentation?</p>
<p>@Robert: Sure, thanks for the idea.  I&#8217;ll get to it next week.</p>
<p>@ant: A kind soul has already written a <a href="http://chadaustin.me/2009/04/flushing-disk-cache/" rel="nofollow">program</a> that does just that.  (It&#8217;s listed on the notes page I linked above.)  It does work, but problem is, it grinds your system to a halt for ten minutes as it does its business.  The more RAM, the longer it takes.  Ts is the average of 20 runs, so that&#8217;s three hours and twenty minutes just to complete one Ts.  Maybe we&#8217;ll have to use it, though.</p>
<p>@Anon #2: Sure, I meant that we would have caught it as we landed it.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Cold Ts by Anonymous</title>
		<link>http://blog.mozilla.com/adw/2009/09/04/cold-ts/#comment-125</link>
		<dc:creator>Anonymous</dc:creator>
		<pubDate>Sat, 05 Sep 2009 23:01:09 +0000</pubDate>
		<guid isPermaLink="false">http://blog.mozilla.com/adw/?p=72#comment-125</guid>
		<description>About the uncaught trunk regression: isn&#039;t it possible to test old builds too, and determine the regression date after the fact?</description>
		<content:encoded><![CDATA[<p>About the uncaught trunk regression: isn&#8217;t it possible to test old builds too, and determine the regression date after the fact?</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Cold Ts by ant</title>
		<link>http://blog.mozilla.com/adw/2009/09/04/cold-ts/#comment-124</link>
		<dc:creator>ant</dc:creator>
		<pubDate>Sat, 05 Sep 2009 17:06:35 +0000</pubDate>
		<guid isPermaLink="false">http://blog.mozilla.com/adw/?p=72#comment-124</guid>
		<description>Seems easy enough - just have a script or something allocate all free RAM plus a few MB extra. That should flush the cache out of memory at least.</description>
		<content:encoded><![CDATA[<p>Seems easy enough &#8211; just have a script or something allocate all free RAM plus a few MB extra. That should flush the cache out of memory at least.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Cold Ts by Robert Kaiser</title>
		<link>http://blog.mozilla.com/adw/2009/09/04/cold-ts/#comment-123</link>
		<dc:creator>Robert Kaiser</dc:creator>
		<pubDate>Sat, 05 Sep 2009 14:19:53 +0000</pubDate>
		<guid isPermaLink="false">http://blog.mozilla.com/adw/?p=72#comment-123</guid>
		<description>I&#039;m sure the prompt with sudo can somehow be worked around, Windows is a problem though (like so often)...

In any case, could you also test Gran Paradiso (3.0) as a comparison so we have another baseline? For hysterical reasons, even a comparison against FF2 might be interesting - I hope we actually did improve somewhere ;-)</description>
		<content:encoded><![CDATA[<p>I&#8217;m sure the prompt with sudo can somehow be worked around, Windows is a problem though (like so often)&#8230;</p>
<p>In any case, could you also test Gran Paradiso (3.0) as a comparison so we have another baseline? For hysterical reasons, even a comparison against FF2 might be interesting &#8211; I hope we actually did improve somewhere ;-)</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Cold Ts by Anonymous</title>
		<link>http://blog.mozilla.com/adw/2009/09/04/cold-ts/#comment-122</link>
		<dc:creator>Anonymous</dc:creator>
		<pubDate>Sat, 05 Sep 2009 07:56:45 +0000</pubDate>
		<guid isPermaLink="false">http://blog.mozilla.com/adw/?p=72#comment-122</guid>
		<description>Allowing sudo for a single command (&quot;sudo /usr/local/sbin/simulate-cold-startup&quot;) shouldn&#039;t introduce a security hole; it would only mean that the user account which runs Talos could decrease system performance.  Seems well worth doing on a test system.</description>
		<content:encoded><![CDATA[<p>Allowing sudo for a single command (&#8221;sudo /usr/local/sbin/simulate-cold-startup&#8221;) shouldn&#8217;t introduce a security hole; it would only mean that the user account which runs Talos could decrease system performance.  Seems well worth doing on a test system.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Cold Ts by adw</title>
		<link>http://blog.mozilla.com/adw/2009/09/04/cold-ts/#comment-121</link>
		<dc:creator>adw</dc:creator>
		<pubDate>Sat, 05 Sep 2009 07:37:03 +0000</pubDate>
		<guid isPermaLink="false">http://blog.mozilla.com/adw/?p=72#comment-121</guid>
		<description>Prasino, Alice and David have recently been working on dirty profile Ts (&lt;a href=&quot;https://bugzilla.mozilla.org/show_bug.cgi?id=414660&quot; rel=&quot;nofollow&quot;&gt;bug 414660&lt;/a&gt; among others), and in fact it&#039;s been on the Tinderboxen since the middle of last month.  If you &lt;a href=&quot;http://tinderbox.mozilla.org/showbuilds.cgi?tree=Firefox&quot; rel=&quot;nofollow&quot;&gt;check them&lt;/a&gt; and search for strings like &quot;ts_places&quot;, you&#039;ll see that large history and bookmarks databases don&#039;t make quite the warm startup impact you might think.  (They really impact shutdown, though.  There is probably a fair amount of low-hanging fruit there.)  Cold startup is a different story and one we have not measured yet.  I think it shouldn&#039;t be too hard to hook up the dirty profile Ts to cold Ts.  Personally I&#039;ve been trying to get a baseline idea of where the problems are and what can be improved without the profile overhead; even with a fresh profile cold startup sucks.  (There&#039;s more to profiles that impacts startup than history by the way -- session restore and add-ons for example.)</description>
		<content:encoded><![CDATA[<p>Prasino, Alice and David have recently been working on dirty profile Ts (<a href="https://bugzilla.mozilla.org/show_bug.cgi?id=414660" rel="nofollow">bug 414660</a> among others), and in fact it&#8217;s been on the Tinderboxen since the middle of last month.  If you <a href="http://tinderbox.mozilla.org/showbuilds.cgi?tree=Firefox" rel="nofollow">check them</a> and search for strings like &#8220;ts_places&#8221;, you&#8217;ll see that large history and bookmarks databases don&#8217;t make quite the warm startup impact you might think.  (They really impact shutdown, though.  There is probably a fair amount of low-hanging fruit there.)  Cold startup is a different story and one we have not measured yet.  I think it shouldn&#8217;t be too hard to hook up the dirty profile Ts to cold Ts.  Personally I&#8217;ve been trying to get a baseline idea of where the problems are and what can be improved without the profile overhead; even with a fresh profile cold startup sucks.  (There&#8217;s more to profiles that impacts startup than history by the way &#8212; session restore and add-ons for example.)</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Cold Ts by Prasino</title>
		<link>http://blog.mozilla.com/adw/2009/09/04/cold-ts/#comment-120</link>
		<dc:creator>Prasino</dc:creator>
		<pubDate>Sat, 05 Sep 2009 07:02:03 +0000</pubDate>
		<guid isPermaLink="false">http://blog.mozilla.com/adw/?p=72#comment-120</guid>
		<description>Can having a large history influence cold start time significantly?  If yes, would it make sense measuring it with a realistic profile too rather than just a clean one?</description>
		<content:encoded><![CDATA[<p>Can having a large history influence cold start time significantly?  If yes, would it make sense measuring it with a realistic profile too rather than just a clean one?</p>
]]></content:encoded>
	</item>
</channel>
</rss>
