BTW, I can reproduce the crash with latest Eliot&#39;s Cog and Pharo vm.<br><br><div class="gmail_quote">On Wed, Feb 20, 2013 at 11:00 AM, Guillermo Polito <span dir="ltr">&lt;<a href="mailto:guillermopolito@gmail.com" target="_blank">guillermopolito@gmail.com</a>&gt;</span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">&#39;cause you don&#39;t know if the method has trailer or not I think... There are empty trailers and trailers embedding source (directly in the image)... isn&#39;t it?<div class="HOEnZb">
<div class="h5"><br><br><div class="gmail_quote">On Wed, Feb 20, 2013 at 10:52 AM, Marcus Denker <span dir="ltr">&lt;<a href="mailto:marcus.denker@inria.fr" target="_blank">marcus.denker@inria.fr</a>&gt;</span> wrote:<br>

<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word"><div><div><br><div><div>On Feb 20, 2013, at 10:45 AM, Guillermo Polito &lt;<a href="mailto:guillermopolito@gmail.com" target="_blank">guillermopolito@gmail.com</a>&gt; wrote:</div>

<br><blockquote type="cite">Hi!<div><br></div><div>There is this bug open I was taking a look yesterday:</div><div><br></div><div><a href="http://code.google.com/p/pharo/issues/detail?can=2&amp;start=0&amp;num=100&amp;q=Milestone%3D2.0&amp;colspec=ID%20Type%20Status%20Summary%20Milestone%20Difficulty&amp;groupby=&amp;sort=&amp;id=7499" target="_blank">http://code.google.com/p/pharo/issues/detail?can=2&amp;start=0&amp;num=100&amp;q=Milestone%3D2.0&amp;colspec=ID%20Type%20Status%20Summary%20Milestone%20Difficulty&amp;groupby=&amp;sort=&amp;id=7499</a></div>


<div><br></div><div>The condense sources compacts the changes and source files, and adds a new method trailer to a compiled method with it&#39;s new source pointer. In order to do that, it does a #becomeForward: to the method.</div>


<div><br></div><div>CompiledMethod&gt;&gt;setSourcePointer: srcPointer</div><div><span style="white-space:pre-wrap">	</span>&quot;We can&#39;t change the trailer of existing method, since</div><div><span style="white-space:pre-wrap">	</span>it could have completely different format. Therefore we need to</div>


<div><span style="white-space:pre-wrap">	</span>generate a copy with new trailer, containing an scrPointer, and then</div><div><span style="white-space:pre-wrap">	</span>#become it&quot;</div>
<div><span style="white-space:pre-wrap">	</span>| trailer copy |</div><div><span style="white-space:pre-wrap">	</span>trailer := CompiledMethodTrailer new sourcePointer: srcPointer.</div>
<div><span style="white-space:pre-wrap">	</span>copy := self copyWithTrailerBytes: trailer.</div><div>&gt;&gt;&gt;<span style="white-space:pre-wrap">	</span>self becomeForward: copy.</div>
<div><span style="white-space:pre-wrap">	</span>^ copy</div><div><br></div><div>So far, with simple methods, so good.</div><div><br></div><div>However, there are cases, in which becoming the compiled method, breaks processes, sometimes getting a crash in the VM. For example:</div>


<div><br></div><div>(Delay class&gt;&gt;#handleTimerEvent) setSourcePosition: 200 inFile: 1</div><div><br></div><div>(ProcessorScheduler class&gt;&gt;#idleProcess) setSourcePosition: 200 inFile: 1</div><div><br></div><div>


My assumption is that the #becomeForward: of compiled methods do not get well with the stack mapping in the vm... But just guessing.</div><div><br></div></blockquote><br></div></div></div><div>I wonder why we don&#39;t just update the pointer.�</div>

<div>That is: we keep the method the same, rewrite the pointer, and nothing else.</div><span><font color="#888888"><div><br></div><div><span style="white-space:pre-wrap">	</span>Marcus</div><br></font></span></div>
</blockquote></div><br>
</div></div></blockquote></div><br>