<?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 on: Dear Lazyweb, am I using memtest86+ correctly?</title>
	<atom:link href="http://strugglers.net/~andy/blog/2007/01/19/dear-lazyweb-am-i-using-memtest86-correctly/feed/" rel="self" type="application/rss+xml" />
	<link>http://strugglers.net/~andy/blog/2007/01/19/dear-lazyweb-am-i-using-memtest86-correctly/</link>
	<description>I'll get there one day.</description>
	<pubDate>Fri, 21 Nov 2008 21:11:44 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.7-beta3</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: L. Bukys</title>
		<link>http://strugglers.net/~andy/blog/2007/01/19/dear-lazyweb-am-i-using-memtest86-correctly/comment-page-1/#comment-8271</link>
		<dc:creator>L. Bukys</dc:creator>
		<pubDate>Tue, 13 Mar 2007 15:27:42 +0000</pubDate>
		<guid isPermaLink="false">http://strugglers.net/~andy/blog/2007/01/19/dear-lazyweb-am-i-using-memtest86-correctly/#comment-8271</guid>
		<description>The BIOS-ALL (e820-All memory config setting) gives spurious error reports.  It does so on every machine I tried.  Then I found this FAQ http://forum.x86-secret.com/showthread.php?t=2807 :

  - When I select BIOS-ALL I get many errors or my machine crashes.

    This is normal. With todays computers this option should never be selected.
    When selecting BIOS-ALL memtest will ignore the e820 memory map supplied by
    the BIOS and will also test areas reserved by the BIOS for the processor or
    the BIOS. See the question about which memory is test for more information
    on this.</description>
		<content:encoded><![CDATA[<p>The BIOS-ALL (e820-All memory config setting) gives spurious error reports.  It does so on every machine I tried.  Then I found this FAQ <a href="http://forum.x86-secret.com/showthread.php?t=2807" rel="nofollow">http://forum.x86-secret.com/showthread.php?t=2807</a> :</p>
<p>  - When I select BIOS-ALL I get many errors or my machine crashes.</p>
<p>    This is normal. With todays computers this option should never be selected.<br />
    When selecting BIOS-ALL memtest will ignore the e820 memory map supplied by<br />
    the BIOS and will also test areas reserved by the BIOS for the processor or<br />
    the BIOS. See the question about which memory is test for more information<br />
    on this.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: L. Bukys</title>
		<link>http://strugglers.net/~andy/blog/2007/01/19/dear-lazyweb-am-i-using-memtest86-correctly/comment-page-1/#comment-8249</link>
		<dc:creator>L. Bukys</dc:creator>
		<pubDate>Tue, 13 Mar 2007 14:51:31 +0000</pubDate>
		<guid isPermaLink="false">http://strugglers.net/~andy/blog/2007/01/19/dear-lazyweb-am-i-using-memtest86-correctly/#comment-8249</guid>
		<description>I had similar behavior on my Dell Inspiron 700m - memory configuration e820-Std never fails, e820-All fails in a block of high memory.  It does make me wonder whether e820-All maps something that always fails, or whether my machine really has bad RAM.  I'm going to take the memtest86 bootable disks to some other machines and see if e820-All always fails on all machines.</description>
		<content:encoded><![CDATA[<p>I had similar behavior on my Dell Inspiron 700m - memory configuration e820-Std never fails, e820-All fails in a block of high memory.  It does make me wonder whether e820-All maps something that always fails, or whether my machine really has bad RAM.  I&#8217;m going to take the memtest86 bootable disks to some other machines and see if e820-All always fails on all machines.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Paul</title>
		<link>http://strugglers.net/~andy/blog/2007/01/19/dear-lazyweb-am-i-using-memtest86-correctly/comment-page-1/#comment-5908</link>
		<dc:creator>Paul</dc:creator>
		<pubDate>Wed, 31 Jan 2007 00:17:23 +0000</pubDate>
		<guid isPermaLink="false">http://strugglers.net/~andy/blog/2007/01/19/dear-lazyweb-am-i-using-memtest86-correctly/#comment-5908</guid>
		<description>Yes, this is normal behavior from my experience of memtest86+. I'm a user, not an expert.

But in every case when I've gotten kernel panics/crashes and ECC errors, memtest86+ has detected a section of bad memory. My vendor doesn't force me to trace it down to the DIMM, so I will just pull pairs of DIMMS and test with memtest86+ a pair at a time until I find the guilty pair, and then swap it for a new pair with the vendor.

This certainly looks like you've got a bad DIMM. 

I don't know how to tell from memtest86+ which physical DIMM it is, so I just do the brute force approach I outlined above. OTOH, depending on your vendor, maybe they'll just send you 4 new DIMMS that you can swap in and they can handle figuring out the bad one.</description>
		<content:encoded><![CDATA[<p>Yes, this is normal behavior from my experience of memtest86+. I&#8217;m a user, not an expert.</p>
<p>But in every case when I&#8217;ve gotten kernel panics/crashes and ECC errors, memtest86+ has detected a section of bad memory. My vendor doesn&#8217;t force me to trace it down to the DIMM, so I will just pull pairs of DIMMS and test with memtest86+ a pair at a time until I find the guilty pair, and then swap it for a new pair with the vendor.</p>
<p>This certainly looks like you&#8217;ve got a bad DIMM. </p>
<p>I don&#8217;t know how to tell from memtest86+ which physical DIMM it is, so I just do the brute force approach I outlined above. OTOH, depending on your vendor, maybe they&#8217;ll just send you 4 new DIMMS that you can swap in and they can handle figuring out the bad one.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: The ongoing struggle &#187; Blog Archive &#187; What Would Lazyweb Do?</title>
		<link>http://strugglers.net/~andy/blog/2007/01/19/dear-lazyweb-am-i-using-memtest86-correctly/comment-page-1/#comment-5411</link>
		<dc:creator>The ongoing struggle &#187; Blog Archive &#187; What Would Lazyweb Do?</dc:creator>
		<pubDate>Sat, 20 Jan 2007 03:27:48 +0000</pubDate>
		<guid isPermaLink="false">http://strugglers.net/~andy/blog/2007/01/19/dear-lazyweb-am-i-using-memtest86-correctly/#comment-5411</guid>
		<description>[...] The ongoing struggle I&#8217;ll get there one day.      &#171; Dear Lazyweb, am I using memtest86+ correctly? [...]</description>
		<content:encoded><![CDATA[<p>[...] The ongoing struggle I&#8217;ll get there one day.      &laquo; Dear Lazyweb, am I using memtest86+ correctly? [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: davee</title>
		<link>http://strugglers.net/~andy/blog/2007/01/19/dear-lazyweb-am-i-using-memtest86-correctly/comment-page-1/#comment-5368</link>
		<dc:creator>davee</dc:creator>
		<pubDate>Fri, 19 Jan 2007 09:14:04 +0000</pubDate>
		<guid isPermaLink="false">http://strugglers.net/~andy/blog/2007/01/19/dear-lazyweb-am-i-using-memtest86-correctly/#comment-5368</guid>
		<description>Not quite answering your question, but thought that I'd just mention that I have used memtest86+ to test the RAM on an amd64 system with 32GB RAM quite happily.  I'd upgraded the RAM from 8GB to 32GB and wanted to give the system a good memtest 'soak' to be happy the new RAM was good.

My test work fine and showed no errors, so it's unlikely you're hitting any kind of memtest/kernel limitation, assuming you're using a recent version of memtest86+.</description>
		<content:encoded><![CDATA[<p>Not quite answering your question, but thought that I&#8217;d just mention that I have used memtest86+ to test the RAM on an amd64 system with 32GB RAM quite happily.  I&#8217;d upgraded the RAM from 8GB to 32GB and wanted to give the system a good memtest &#8217;soak&#8217; to be happy the new RAM was good.</p>
<p>My test work fine and showed no errors, so it&#8217;s unlikely you&#8217;re hitting any kind of memtest/kernel limitation, assuming you&#8217;re using a recent version of memtest86+.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
