<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
<br>
Hi Richard,<br>
<br>
I have just committed a fix to this problem. It was due to a
lack of checking whether an internal pointer in the
DReferenceTrajectory object was set prior to using it. This pointer
is set by calls to certain methods (like DistToRT()) but then used
when one wants additional information regarding the point on the RT
that was determined in the previous call. If the previous call fails
to find the POCA for whatever reason, it sets this internal pointer
to NULL indicating to subsequent calls to GetLastDOCAPoint methods
(of which there are two) that they should consider this information
invalid.<br>
<br>
Could you please verify that this fixes the problem for you?<br>
<br>
Rgards,<br>
-David<br>
<br>
On 3/21/11 9:32 AM, Richard Jones wrote:
<blockquote cite="mid:4D87538A.6030303@uconn.edu" type="cite">
David,<br>
<br>
Just checking -- "not YET gotten a segfault" ?<br>
<br>
Do you mean, do not get a segfault? This file contains only one
event, so either you do get one when you analyze it or you don't.
If you don't, this is a clue. So far I have reproduced it under
the following circumstances:<br>
<ol>
<li>default build, i686 without sse<br>
</li>
<li>default build, x86_64 with sse2, sse3</li>
<li>debug build with OPTIMIZE=2, x86_64 with sse2,sse3</li>
</ol>
Next I will try debug build with OPTIMIZE=0<br>
<br>
-Richard J.<br>
<br>
<br>
On 3/21/2011 9:28 AM, David Lawrence wrote:
<blockquote type="cite" cite="mid:4D875265.7000506@jlab.org">
<pre wrap="">Hi Richard,
I've tried reproducing the error on Mac OS X 10.6 using revision 7602 but have not yet gotten a seg. fault. I'm in the process of compiling the same version on CentOS 5.3 x86_64 to see if I can reproduce the problem there.
Regards,
-David
On 3/21/11 7:51 AM, Richard Jones wrote:
Colleagues,
I have entered this latest issue in Mantis as issue #51. See that for details. This message is just a heads-up to those doing reconstruction studies that this issue is known and reproducible on a data file containing a single event. A fix should be forthcoming shortly.
-Richard J.
_______________________________________________
Halld-offline mailing list
<a moz-do-not-send="true" href="mailto:Halld-offline@jlab.org">Halld-offline@jlab.org</a><a moz-do-not-send="true" href="mailto:Halld-offline@jlab.org"><mailto:Halld-offline@jlab.org></a>
<a moz-do-not-send="true" href="https://mailman.jlab.org/mailman/listinfo/halld-offline">https://mailman.jlab.org/mailman/listinfo/halld-offline</a>
</pre>
</blockquote>
<br>
</blockquote>
</body>
</html>