<?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 Persistent notes</title>
	<atom:link href="http://alexkr.com/comments/feed/" rel="self" type="application/rss+xml" />
	<link>http://alexkr.com</link>
	<description>Alexander Krivutsenko&#039;s online journal</description>
	<lastBuildDate>Tue, 24 Aug 2010 16:35:21 +0000</lastBuildDate>
	<generator>http://wordpress.org/?v=2.9.2</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>Comment on Layered windows and UpdateLayeredWindow by Eric Layne</title>
		<link>http://alexkr.com/source-code/50/layered-windows-and-updatelayeredwindow/comment-page-1/#comment-39279</link>
		<dc:creator>Eric Layne</dc:creator>
		<pubDate>Tue, 24 Aug 2010 16:35:21 +0000</pubDate>
		<guid isPermaLink="false">http://alexkr.com/windows/50/layered-windows-and-updatelayeredwindow/#comment-39279</guid>
		<description>Good article, thanks for posting this information.</description>
		<content:encoded><![CDATA[<p>Good article, thanks for posting this information.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Current state of SCTP by Bruce Cran</title>
		<link>http://alexkr.com/memos/156/sctp-on-windows/comment-page-1/#comment-36339</link>
		<dc:creator>Bruce Cran</dc:creator>
		<pubDate>Wed, 30 Jun 2010 19:04:56 +0000</pubDate>
		<guid isPermaLink="false">http://alexkr.com/memos/156/using-sctp-for-better-reliability/#comment-36339</guid>
		<description>The version of sctpDrv available via the co-conv site appears to have been unmaintained since the end of 2008. I took the code that was made available via sctp.jp and have been fixing bugs and making improvements such as adding 64-bit support and fixing installation issues. The new SctpDrv site is at http://www.bluestop.org/SctpDrv/ .</description>
		<content:encoded><![CDATA[<p>The version of sctpDrv available via the co-conv site appears to have been unmaintained since the end of 2008. I took the code that was made available via sctp.jp and have been fixing bugs and making improvements such as adding 64-bit support and fixing installation issues. The new SctpDrv site is at <a href="http://www.bluestop.org/SctpDrv/" rel="nofollow">http://www.bluestop.org/SctpDrv/</a> .</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on The IP slash notation by Masking an IP &#124; Hide Your IP Address</title>
		<link>http://alexkr.com/memos/345/ip-slash-notation/comment-page-1/#comment-30904</link>
		<dc:creator>Masking an IP &#124; Hide Your IP Address</dc:creator>
		<pubDate>Tue, 30 Mar 2010 22:02:01 +0000</pubDate>
		<guid isPermaLink="false">http://alexkr.com/?p=345#comment-30904</guid>
		<description>[...] The IP slash notation « Persistent notes [...]</description>
		<content:encoded><![CDATA[<p>[...] The IP slash notation « Persistent notes [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on How to use hash_map by thomas</title>
		<link>http://alexkr.com/source-code/13/how-to-use-hash_map/comment-page-1/#comment-21570</link>
		<dc:creator>thomas</dc:creator>
		<pubDate>Wed, 11 Nov 2009 19:17:32 +0000</pubDate>
		<guid isPermaLink="false">http://alexkr.com/uncategorized/13/how-to-use-hash_map/#comment-21570</guid>
		<description>oops, it butchered my comment because it had less than and greater than signs.. basically you need to pass a fourth thing to the hash_map template, which is the same as the third argument</description>
		<content:encoded><![CDATA[<p>oops, it butchered my comment because it had less than and greater than signs.. basically you need to pass a fourth thing to the hash_map template, which is the same as the third argument</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on How to use hash_map by thomas</title>
		<link>http://alexkr.com/source-code/13/how-to-use-hash_map/comment-page-1/#comment-21569</link>
		<dc:creator>thomas</dc:creator>
		<pubDate>Wed, 11 Nov 2009 19:16:15 +0000</pubDate>
		<guid isPermaLink="false">http://alexkr.com/uncategorized/13/how-to-use-hash_map/#comment-21569</guid>
		<description>typedef hash_map &gt; HashObject;

should be:

typedef hash_map, HashConfiguration &gt; HashObject;

(or at least that&#039;s what I found when I tried this!)</description>
		<content:encoded><![CDATA[<p>typedef hash_map &gt; HashObject;</p>
<p>should be:</p>
<p>typedef hash_map, HashConfiguration &gt; HashObject;</p>
<p>(or at least that&#8217;s what I found when I tried this!)</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Detecting TCP connection problems by Dave</title>
		<link>http://alexkr.com/posts/150/tcp-keep-alives/comment-page-1/#comment-16430</link>
		<dc:creator>Dave</dc:creator>
		<pubDate>Sat, 12 Sep 2009 18:25:49 +0000</pubDate>
		<guid isPermaLink="false">http://alexkr.com/posts/150/tcp-keep-alives/#comment-16430</guid>
		<description>Okay, this is awesome. I had no idea it existed. I&#039;ve never had a need to prevent disconnection due to inactivity, but as you know, I am always on the lookout for more robust ways of detecting when my peer has died without properly closing its TCP connection.

--Dave</description>
		<content:encoded><![CDATA[<p>Okay, this is awesome. I had no idea it existed. I&#8217;ve never had a need to prevent disconnection due to inactivity, but as you know, I am always on the lookout for more robust ways of detecting when my peer has died without properly closing its TCP connection.</p>
<p>&#8211;Dave</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Non blocking singly linked list by Alex</title>
		<link>http://alexkr.com/memos/16/non-blocking-singly-linked-list/comment-page-1/#comment-2556</link>
		<dc:creator>Alex</dc:creator>
		<pubDate>Sat, 18 Oct 2008 18:33:34 +0000</pubDate>
		<guid isPermaLink="false">http://alexkr.com/windows/16/non-blocking-singly-linked-list/#comment-2556</guid>
		<description>ChasM, thank you for your useful comment.

MSDN indeed says:

All list items must be aligned on a MEMORY_ALLOCATION_ALIGNMENT boundary; otherwise, this function will behave unpredictably.

Instead of proper allocation (which will cause editing this post) I may add some special requirements for the logging API.
Considering that on Win32 MEMORY_ALLOCATION_ALIGNMENT is 8 that makes my example suitable only for strings with length equal to 3, 11, 19,..  etc :) 
</description>
		<content:encoded><![CDATA[<p>ChasM, thank you for your useful comment.</p>
<p>MSDN indeed says:</p>
<p>All list items must be aligned on a MEMORY_ALLOCATION_ALIGNMENT boundary; otherwise, this function will behave unpredictably.</p>
<p>Instead of proper allocation (which will cause editing this post) I may add some special requirements for the logging API.<br />
Considering that on Win32 MEMORY_ALLOCATION_ALIGNMENT is 8 that makes my example suitable only for strings with length equal to 3, 11, 19,..  etc :)</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on Non blocking singly linked list by ChasM</title>
		<link>http://alexkr.com/memos/16/non-blocking-singly-linked-list/comment-page-1/#comment-2530</link>
		<dc:creator>ChasM</dc:creator>
		<pubDate>Fri, 17 Oct 2008 16:11:25 +0000</pubDate>
		<guid isPermaLink="false">http://alexkr.com/windows/16/non-blocking-singly-linked-list/#comment-2530</guid>
		<description>Comments on WriteLogMessage:
I think the way you&#039;ve tacked storage for the strText pointer as well as the string itself is pretty clever. I wouldn&#039;t have thought of that.

And although it&#039;s clearer to use new and delete like you did, the SList Interlocked functions require each allocated block to be aligned on a dword boundary, which new doesn&#039;t do. You need to use _aligned_malloc and _aligned_free instead. The requirement for alignment is imposed by the CPU. This is a gotcha that would be very hard to debug....</description>
		<content:encoded><![CDATA[<p>Comments on WriteLogMessage:<br />
I think the way you&#8217;ve tacked storage for the strText pointer as well as the string itself is pretty clever. I wouldn&#8217;t have thought of that.</p>
<p>And although it&#8217;s clearer to use new and delete like you did, the SList Interlocked functions require each allocated block to be aligned on a dword boundary, which new doesn&#8217;t do. You need to use _aligned_malloc and _aligned_free instead. The requirement for alignment is imposed by the CPU. This is a gotcha that would be very hard to debug&#8230;.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
