<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
	<channel>
		<title><![CDATA[Approximatrix Forums — Warning: expressly disable all optimisation when debugging]]></title>
		<link>https://forums.approximatrix.com/viewtopic.php?id=312</link>
		<atom:link href="https://forums.approximatrix.com/extern.php?action=feed&amp;tid=312&amp;type=rss" rel="self" type="application/rss+xml" />
		<description><![CDATA[The most recent posts in Warning: expressly disable all optimisation when debugging.]]></description>
		<lastBuildDate>Tue, 04 Feb 2014 15:03:01 +0000</lastBuildDate>
		<generator>PunBB</generator>
		<item>
			<title><![CDATA[Warning: expressly disable all optimisation when debugging]]></title>
			<link>https://forums.approximatrix.com/viewtopic.php?pid=1254#p1254</link>
			<description><![CDATA[<p>Fellow users who haven&#039;t encountered this before, as I hadn&#039;t prior to SF2.10 (which <em>I think</em> introduces a code optimization selector that defaults to &#039;Common&#039;) might appreciate this warning.</p><p>When selecting &#039;Enable Debugging&#039;, in </p><p>&nbsp; &nbsp;<strong>Options &gt; Project &gt; Code generation</strong> ,</p><p>it will generally be important to ensure that </p><p>&nbsp; &nbsp;<strong>Aggressive loop optimizations</strong> is <span class="bbu">un-ticked</span>,<br />&nbsp; &nbsp;<strong>Optimization level</strong> is set to <span class="bbu">none</span>.</p><p>This is because, with optimized code, two things will happen in debugging sessions which can make debugging very difficult.&nbsp; One is that the debugger will not work through the program program source code line by line, just as the code is written, but will leap about in seemingly-incomprehensible ways.&nbsp; The other is that many variables, which FORTRAN developers might need to watch during debugging sessions, will be reported as &#039;Optimized out&#039;.</p><p>I&#039;ve suggested that it might be best for enabling debugging automatically to kill all optimizing.&nbsp; Jeff is considering it.<br />--- <br />John</p>]]></description>
			<author><![CDATA[null@example.com (JohnWasilewski)]]></author>
			<pubDate>Tue, 04 Feb 2014 15:03:01 +0000</pubDate>
			<guid>https://forums.approximatrix.com/viewtopic.php?pid=1254#p1254</guid>
		</item>
	</channel>
</rss>
