<?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: Are we having, like, a conversation?</title>
	<atom:link href="http://digitalvampire.org/blog/index.php/2007/06/07/are-we-having-like-a-conversation/feed/" rel="self" type="application/rss+xml" />
	<link>http://digitalvampire.org/blog/index.php/2007/06/07/are-we-having-like-a-conversation/</link>
	<description>Linux hacker, recovering mathematician, former athlete</description>
	<lastBuildDate>Fri, 11 Mar 2011 20:17:40 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.1</generator>
	<item>
		<title>By: roland</title>
		<link>http://digitalvampire.org/blog/index.php/2007/06/07/are-we-having-like-a-conversation/comment-page-1/#comment-91</link>
		<dc:creator>roland</dc:creator>
		<pubDate>Fri, 08 Jun 2007 04:06:49 +0000</pubDate>
		<guid isPermaLink="false">http://digitalvampire.org/blog/index.php/2007/06/07/are-we-having-like-a-conversation/#comment-91</guid>
		<description>oh, my bad.  IO ports are so old-school I just read outw/inw as writew/readw without thinking about it -- I never looked at what the UHCI interface looked like before, but sure enough all the PCI device has is 32 bytes of IO ports.

But surely outw() and inw() are also defined to be ordered with respect to each other?  An mb() between outw() and inw() must be totally redundant, or else code that poked at ports in a specific order would be totally insane.  So I still think the code is at least somewhat ugly.</description>
		<content:encoded><![CDATA[<p>oh, my bad.  IO ports are so old-school I just read outw/inw as writew/readw without thinking about it &#8212; I never looked at what the UHCI interface looked like before, but sure enough all the PCI device has is 32 bytes of IO ports.</p>
<p>But surely outw() and inw() are also defined to be ordered with respect to each other?  An mb() between outw() and inw() must be totally redundant, or else code that poked at ports in a specific order would be totally insane.  So I still think the code is at least somewhat ugly.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: kyle mcmartin</title>
		<link>http://digitalvampire.org/blog/index.php/2007/06/07/are-we-having-like-a-conversation/comment-page-1/#comment-90</link>
		<dc:creator>kyle mcmartin</dc:creator>
		<pubDate>Fri, 08 Jun 2007 03:57:14 +0000</pubDate>
		<guid isPermaLink="false">http://digitalvampire.org/blog/index.php/2007/06/07/are-we-having-like-a-conversation/#comment-90</guid>
		<description>ioports (using in/out) are defined to have non-postable semantics, so the mb() is really just a compiler barrier in your usb case, afaict.</description>
		<content:encoded><![CDATA[<p>ioports (using in/out) are defined to have non-postable semantics, so the mb() is really just a compiler barrier in your usb case, afaict.</p>
]]></content:encoded>
	</item>
</channel>
</rss>

