<?xml version="1.0" encoding="iso-8859-1"?><!-- generator="b2evolution/2.4.5" -->
<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:admin="http://webns.net/mvcb/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:content="http://purl.org/rss/1.0/modules/content/">
	<channel>
		<title>John Mylchreest - Latest comments on Kernel Sources</title>
		<link>http://blogs.gentoo.org/johnm?disp=comments</link>
		<description></description>
		<language>en-US</language>
		<docs>http://backend.userland.com/rss</docs>
		<admin:generatorAgent rdf:resource="http://b2evolution.net/?v=2.4.5"/>
		<ttl>60</ttl>
				<item>
			<title>In response to: Kernel Sources</title>
			<pubDate>Thu, 17 Nov 2005 20:40:32 +0000</pubDate>
			<dc:creator>nicolas [Visitor]</dc:creator>
			<guid isPermaLink="false">c1645@http://blogs.gentoo.org/</guid>
			<description>You don't know kernel 32??</description>
			<content:encoded><![CDATA[You don't know kernel 32??]]></content:encoded>
			<link>http://blogs.gentoo.org/johnm/2005/04/05/kernel_sources#c1645</link>
		</item>
				<item>
			<title>In response to: Kernel Sources</title>
			<pubDate>Wed, 07 Sep 2005 17:42:44 +0000</pubDate>
			<dc:creator>Greg [Visitor]</dc:creator>
			<guid isPermaLink="false">c838@http://blogs.gentoo.org/</guid>
			<description>What about Kernel 32 ?</description>
			<content:encoded><![CDATA[What about Kernel 32 ?]]></content:encoded>
			<link>http://blogs.gentoo.org/johnm/2005/04/05/kernel_sources#c838</link>
		</item>
				<item>
			<title>In response to: Kernel Sources</title>
			<pubDate>Wed, 31 Aug 2005 22:16:50 +0000</pubDate>
			<dc:creator>will [Visitor]</dc:creator>
			<guid isPermaLink="false">c825@http://blogs.gentoo.org/</guid>
			<description>okaaaay ;)</description>
			<content:encoded><![CDATA[okaaaay ;)]]></content:encoded>
			<link>http://blogs.gentoo.org/johnm/2005/04/05/kernel_sources#c825</link>
		</item>
				<item>
			<title>In response to: Kernel Sources</title>
			<pubDate>Wed, 13 Apr 2005 13:32:31 +0000</pubDate>
			<dc:creator>HIGH-FREQ [Visitor]</dc:creator>
			<guid isPermaLink="false">c54@http://blogs.gentoo.org/</guid>
			<description>I was stumped on migrating from 2.4 to 2.6 headers as i've read this site up and down..but then again i noticed johnm wrot:&lt;br /&gt;
&lt;br /&gt;
&quot;If you find that its installing a version you dont want, then just relink your /etc/make.profile to ${PORTDIR}/profiles/default-linux/x86/2005.0/XX where XX is 2.4 (or 2.6 on different archs in some cases).&quot;&lt;br /&gt;
&lt;br /&gt;
which that is mistaken due to no 2.6 being listed in that dir.  I then spoke with johnm and its default for 2.6 headers is just symlinking /etc/make.profile to ${PORTDIR}/profiles/default-linux/x86/2005.0/   </description>
			<content:encoded><![CDATA[I was stumped on migrating from 2.4 to 2.6 headers as i've read this site up and down..but then again i noticed johnm wrot:<br />
<br />
"If you find that its installing a version you dont want, then just relink your /etc/make.profile to ${PORTDIR}/profiles/default-linux/x86/2005.0/XX where XX is 2.4 (or 2.6 on different archs in some cases)."<br />
<br />
which that is mistaken due to no 2.6 being listed in that dir.  I then spoke with johnm and its default for 2.6 headers is just symlinking /etc/make.profile to ${PORTDIR}/profiles/default-linux/x86/2005.0/   ]]></content:encoded>
			<link>http://blogs.gentoo.org/johnm/2005/04/05/kernel_sources#c54</link>
		</item>
				<item>
			<title>In response to: Kernel Sources</title>
			<pubDate>Tue, 12 Apr 2005 16:55:55 +0000</pubDate>
			<dc:creator>John Mylchreest [Member]</dc:creator>
			<guid isPermaLink="false">c52@http://blogs.gentoo.org/</guid>
			<description>interesting concept actually that you recommend using einfo's as a way to put forward changes. However, would this not be too late? lets face it, the kernel headers really didnt change for most people, and by the time they did, it would have been too late to read since you just got upgraded to 2.6?&lt;br /&gt;
&lt;br /&gt;
Or perhaps I'm missing what you mean. Anyways, this move was due, and thankfully its all turned out remarkably well.</description>
			<content:encoded><![CDATA[interesting concept actually that you recommend using einfo's as a way to put forward changes. However, would this not be too late? lets face it, the kernel headers really didnt change for most people, and by the time they did, it would have been too late to read since you just got upgraded to 2.6?<br />
<br />
Or perhaps I'm missing what you mean. Anyways, this move was due, and thankfully its all turned out remarkably well.]]></content:encoded>
			<link>http://blogs.gentoo.org/johnm/2005/04/05/kernel_sources#c52</link>
		</item>
				<item>
			<title>In response to: Kernel Sources</title>
			<pubDate>Tue, 12 Apr 2005 15:49:53 +0000</pubDate>
			<dc:creator>Tracey Clark [Visitor]</dc:creator>
			<guid isPermaLink="false">c51@http://blogs.gentoo.org/</guid>
			<description>Overall, I'm glad Gentoo has made the 2.6 kernel the default, while providing a way to keep 2.4 kernels around. Gentoo continues to be my distro of choice for many reasons. I have to agree, however, that notification of the header conflict could have been better advertised beforehand.&lt;br /&gt;
&lt;br /&gt;
One way to provide advanced warning for upcoming changes such as with linux-headers is to write a message that is displayed before or after an emerge. This has been done with other packages. It is much more likely to be seen, and with direct relevance to the issue, than putting the information in the weekly newsletter. As good as it is, not everyone reads the GWN. </description>
			<content:encoded><![CDATA[Overall, I'm glad Gentoo has made the 2.6 kernel the default, while providing a way to keep 2.4 kernels around. Gentoo continues to be my distro of choice for many reasons. I have to agree, however, that notification of the header conflict could have been better advertised beforehand.<br />
<br />
One way to provide advanced warning for upcoming changes such as with linux-headers is to write a message that is displayed before or after an emerge. This has been done with other packages. It is much more likely to be seen, and with direct relevance to the issue, than putting the information in the weekly newsletter. As good as it is, not everyone reads the GWN. ]]></content:encoded>
			<link>http://blogs.gentoo.org/johnm/2005/04/05/kernel_sources#c51</link>
		</item>
				<item>
			<title>In response to: Kernel Sources</title>
			<pubDate>Tue, 12 Apr 2005 12:34:56 +0000</pubDate>
			<dc:creator>John Mylchreest [Member]</dc:creator>
			<guid isPermaLink="false">c49@http://blogs.gentoo.org/</guid>
			<description>Hi TJ, thanks for the feedback.&lt;br /&gt;
Just to run by a few of the above comments.&lt;br /&gt;
&lt;br /&gt;
Of course I understand your point of view, however for 2 issues in the GWN before the release, we stated how important it was to change to a cascading profile upon the release date to avoid this mass upgrade. We tried to advertise as loudly as possible about the migration documents and so on.&lt;br /&gt;
&lt;br /&gt;
keeping linux26-headers and linux-headers was fundementally wrong, as stated above since this becomes very problematic from a porage point of view.&lt;br /&gt;
&lt;br /&gt;
This was one of the most radical changes we (the Kernel Team) have had to make in quite some time. If you could donate some examples of how we might be able to improve this kind of announcement in future then they will be welcome.</description>
			<content:encoded><![CDATA[Hi TJ, thanks for the feedback.<br />
Just to run by a few of the above comments.<br />
<br />
Of course I understand your point of view, however for 2 issues in the GWN before the release, we stated how important it was to change to a cascading profile upon the release date to avoid this mass upgrade. We tried to advertise as loudly as possible about the migration documents and so on.<br />
<br />
keeping linux26-headers and linux-headers was fundementally wrong, as stated above since this becomes very problematic from a porage point of view.<br />
<br />
This was one of the most radical changes we (the Kernel Team) have had to make in quite some time. If you could donate some examples of how we might be able to improve this kind of announcement in future then they will be welcome.]]></content:encoded>
			<link>http://blogs.gentoo.org/johnm/2005/04/05/kernel_sources#c49</link>
		</item>
				<item>
			<title>In response to: Kernel Sources</title>
			<pubDate>Tue, 12 Apr 2005 01:47:22 +0000</pubDate>
			<dc:creator>tj [Visitor]</dc:creator>
			<guid isPermaLink="false">c47@http://blogs.gentoo.org/</guid>
			<description>I think the move was due and I don't discuss that.  What I deplore is the cowboy way that this update has been done.  By simply adding a 2.6 version to the linux-header group, it provoked an update for those who had been using the 2.4 version without using spcific cascaded profile.&lt;br /&gt;
&lt;br /&gt;
I run a mix of 2.4 and 2.6.  On the Openmosix cluster nodes, it did not hit me that the linux-headers would be upgraded from 2.4 to 2.6.  I then had segfaults on subsequent emerges.&lt;br /&gt;
&lt;br /&gt;
All in all, I had to manage this to find what went wrong, learn how to fix it, and reverse the damages.  Lost of time, lost of confidence toward the Gentoo sumities that take those kinds of decisions.  Sometimes, introductions of changes are managed with what seems to be (in my view) amateurism.  It's my personal opinion.</description>
			<content:encoded><![CDATA[I think the move was due and I don't discuss that.  What I deplore is the cowboy way that this update has been done.  By simply adding a 2.6 version to the linux-header group, it provoked an update for those who had been using the 2.4 version without using spcific cascaded profile.<br />
<br />
I run a mix of 2.4 and 2.6.  On the Openmosix cluster nodes, it did not hit me that the linux-headers would be upgraded from 2.4 to 2.6.  I then had segfaults on subsequent emerges.<br />
<br />
All in all, I had to manage this to find what went wrong, learn how to fix it, and reverse the damages.  Lost of time, lost of confidence toward the Gentoo sumities that take those kinds of decisions.  Sometimes, introductions of changes are managed with what seems to be (in my view) amateurism.  It's my personal opinion.]]></content:encoded>
			<link>http://blogs.gentoo.org/johnm/2005/04/05/kernel_sources#c47</link>
		</item>
			</channel>
</rss>
