<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><div>The difference is for:</div><div><br></div><div>Event #1339, Track Candidate #2, DTrackCandidate pz & z0.</div><div><br></div><div>We need to find which track candidate factory this is coming from still.  </div><div><br></div><div> - Paul</div><br><div><div>On Mar 14, 2014, at 8:44 AM, Beni Zihlmann wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite">
  
    <meta content="text/html; charset=ISO-8859-1" http-equiv="Content-Type">
  
  <div bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">Hi Paul,<br>
      yea! I see a very similar thing. I run single threaded but with
      two different groups<br>
      once using hd_root and once using hd_dump both with my plugin. The
      group<br>
      using hd_root is consistent in itself and the group with hd_dump
      agrees with<br>
      the group of hd_root except for one run! And yes its event 1339!<br>
      <br>
      Event  #Tracks #CDCHits #FDCHits #+tracsk #-tracks #Neutrals<br>
---------------------------------------------------------------------------------------<br>
      1336    8     51  169  6  2   11<br>
      1337    9    68  491  7  2   7<br>
      1338    16  81  817  10  6   6<br>
      1339    21  43  1398  11  10   17     >>>>> 1339   
      21  43  1398  11  10   18<br>
      1340    14  42  1067  8  6   16<br>
      <br>
      <br>
      different # of Neutrals!<br>
      <br>
      cheers,<br>
      Beni<br>
      <br>
    </div>
    <blockquote cite="mid:C9201787-1143-40F8-942B-D2BEE9347F60@jlab.org" type="cite">
      <div>I ran 32 single-threaded jobs with the new software, and I
        see nearly identical results, but not quite.  Half of the REST
        files have one identical file size (and identical contents), and
        the other half have a different identical file size (and
        identical contents).  This is true whether I run with saving the
        time-based tracking results or the wire-based tracking results
        to REST.  The attached "diff" files show the difference for each
        case (via hddm-xml).  </div>
      <div><br>
      </div>
      <div>This difference is isolated to one event: #1339 in the file I
        linked everyone to a week ago.  In fact, while three wire-based
        and time-based tracks are listed, they all have the same
        candidate id: 2.  </div>
      <div><br>
      </div>
      <div>We've got it cornered Simon!  Now let's finish it off! </div>
      <div><br>
      </div>
      <div> - Paul</div>
      <div><br>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <div><br>
      </div>
      <div><br>
      </div>
      <div><br>
      </div>
      <div>
        <div>On Mar 13, 2014, at 6:56 PM, Sean Dobbs wrote:</div>
        <br class="Apple-interchange-newline">
        <blockquote type="cite">
          <div dir="ltr">
            <div class="gmail_quote"><br>
              <br>
              <div dir="ltr">Hi all,
                <div><br>
                </div>
                <div>I checked out and built a clean version with the
                  new tags and am now seeing consistent results when
                  running with one thread and 4 threads.</div>
                <span class="HOEnZb"><font color="#888888">
                    <div><br>
                    </div>
                    <div><br>
                    </div>
                    <div>---Sean</div>
                  </font></span></div>
              <div class="HOEnZb">
                <div class="h5">
                  <div class="gmail_extra"><br>
                    <br>
                    <div class="gmail_quote">On Thu, Mar 13, 2014 at
                      2:37 PM, Mark Ito <span dir="ltr"><<a moz-do-not-send="true" href="mailto:marki@jlab.org" target="_blank">marki@jlab.org</a>></span>
                      wrote:<br>
                      <blockquote class="gmail_quote" style="margin:0 0
                        0 .8ex;border-left:1px #ccc
                        solid;padding-left:1ex">
                        Still seeing differences of the same ilk as
                        previously reported.<br>
                        <div>
                          <div><br>
                            On 03/13/2014 02:31 PM, Mark Ito wrote:<br>
                            > I've re-tagged to reflect this change:
                            tags/sim-recon-2.5 .<br>
                            ><br>
                            > On 03/13/2014 02:12 PM, Simon Taylor
                            wrote:<br>
                            >> I have checked in some changes to
                            the tracking code that appear to<br>
                            >> address the valgrind errors
                            mentioned below.<br>
                            >><br>
                            >> Simon<br>
                            >><br>
                            >> On 03/12/2014 03:59 PM, Matthew
                            Shepherd wrote:<br>
                            >>> Having just spent many
                            frustrating hours hunting down my own
                            separate non-deterministic bug I was
                            motivated to run hd_dump -DTrackWireBased
                            through valgrind.<br>
                            >>><br>
                            >>> The error below seems
                            suspicious and could result in
                            non-determinstic behaviour, although
                            valgrind is known to generate "errors" where
                            there are none.  I didn't have time to look
                            at the code since I have to run to another
                            meeting, but thought I would pass it on.<br>
                            >>><br>
                            >>> Matt<br>
                            >>><br>
                            >>><br>
                            >>> ==7443== Conditional jump or
                            move depends on uninitialised value(s)<br>
                            >>> ==7443==    at 0x88F769:
                            DTrackFitterKalmanSIMD::KalmanForwardCDC(double,
                            DMatrix5x1&, DMatrix5x5&,
                            double&, unsigned int&) (in
/home/fs1/mashephe/gluex/my_src/bin/Linux_CentOS6-x86_64-gcc4.4.6/hd_dump)<br>
                            >>> ==7443==    by 0x8926C4:
                            DTrackFitterKalmanSIMD::ForwardCDCFit(DMatrix5x1
                            const&, DMatrix5x5 const&) (in
/home/fs1/mashephe/gluex/my_src/bin/Linux_CentOS6-x86_64-gcc4.4.6/hd_dump)<br>
                            >>> ==7443==    by 0x89762A:
                            DTrackFitterKalmanSIMD::KalmanLoop() (in
/home/fs1/mashephe/gluex/my_src/bin/Linux_CentOS6-x86_64-gcc4.4.6/hd_dump)<br>
                            >>> ==7443==    by 0x898315:
                            DTrackFitterKalmanSIMD::FitTrack() (in
/home/fs1/mashephe/gluex/my_src/bin/Linux_CentOS6-x86_64-gcc4.4.6/hd_dump)<br>
                            >>> ==7443==    by 0x84B8CB:
                            DTrackFitter::FindHitsAndFitTrack(DKinematicData
                            const&, DReferenceTrajectory const*,
                            jana::JEventLoop*, double, int, double,
                            DetectorSystem_t) (in
/home/fs1/mashephe/gluex/my_src/bin/Linux_CentOS6-x86_64-gcc4.4.6/hd_dump)<br>
                            >>> ==7443==    by 0x8C25D7:
                            DTrackWireBased_factory::DoFit(unsigned int,
                            DTrackCandidate const*,
                            DReferenceTrajectory*, jana::JEventLoop*,
                            double) (in
/home/fs1/mashephe/gluex/my_src/bin/Linux_CentOS6-x86_64-gcc4.4.6/hd_dump)<br>
                            >>> ==7443==    by 0x8C4803:
                            DTrackWireBased_factory::evnt(jana::JEventLoop*,
                            int) (in
/home/fs1/mashephe/gluex/my_src/bin/Linux_CentOS6-x86_64-gcc4.4.6/hd_dump)<br>
                            >>> ==7443==    by 0x6C8F38:
                            jana::JFactory<DTrackWireBased>::Get(std::vector<DTrackWireBased
                            const*, std::allocator<DTrackWireBased
                            const*> >&) (in
/home/fs1/mashephe/gluex/my_src/bin/Linux_CentOS6-x86_64-gcc4.4.6/hd_dump)<br>
                            >>> ==7443==    by 0x6C97EC:
                            jana::JFactory<DTrackWireBased>*
                            jana::JEventLoop::GetFromFactory<DTrackWireBased>(std::vector<DTrackWireBased
                            const*, std::allocator<DTrackWireBased
                            const*> >&, char const*,
                            jana::JEventLoop::data_source_t&, bool)
                            (in
/home/fs1/mashephe/gluex/my_src/bin/Linux_CentOS6-x86_64-gcc4.4.6/hd_dump)<br>
                            >>> ==7443==    by 0x6C9A84:
                            jana::JFactory<DTrackWireBased>*
                            jana::JEventLoop::Get<DTrackWireBased>(std::vector<DTrackWireBased
                            const*, std::allocator<DTrackWireBased
                            const*> >&, char const*, bool) (in
/home/fs1/mashephe/gluex/my_src/bin/Linux_CentOS6-x86_64-gcc4.4.6/hd_dump)<br>
                            >>> ==7443==    by 0x6CA153:
                            jana::JFactory<DTrackWireBased>::GetNrows()
                            (in
/home/fs1/mashephe/gluex/my_src/bin/Linux_CentOS6-x86_64-gcc4.4.6/hd_dump)<br>
                            >>> ==7443==    by 0x571A39:
                            MyProcessor::evnt(jana::JEventLoop*, int)
                            (in
/home/fs1/mashephe/gluex/my_src/bin/Linux_CentOS6-x86_64-gcc4.4.6/hd_dump)<br>
                            >>><br>
                            >>><br>
                            >>
                            _______________________________________________<br>
                            >> Halld-offline mailing list<br>
                            >> <a moz-do-not-send="true" href="mailto:Halld-offline@jlab.org" target="_blank">Halld-offline@jlab.org</a><br>
                            >> <a moz-do-not-send="true" href="https://mailman.jlab.org/mailman/listinfo/halld-offline" target="_blank">https://mailman.jlab.org/mailman/listinfo/halld-offline</a><br>
                            <br>
                            --<br>
                            Mark M. Ito, Jefferson Lab, <a moz-do-not-send="true" href="mailto:marki@jlab.org" target="_blank">marki@jlab.org</a>, <a moz-do-not-send="true" href="tel:%28757%29269-5295" value="+17572695295" target="_blank">(757)269-5295</a><br>
                            <br>
_______________________________________________<br>
                            Halld-offline mailing list<br>
                            <a moz-do-not-send="true" href="mailto:Halld-offline@jlab.org" target="_blank">Halld-offline@jlab.org</a><br>
                            <a moz-do-not-send="true" href="https://mailman.jlab.org/mailman/listinfo/halld-offline" target="_blank">https://mailman.jlab.org/mailman/listinfo/halld-offline</a><br>
                          </div>
                        </div>
                      </blockquote>
                    </div>
                    <br>
                  </div>
                </div>
              </div>
            </div>
            <br>
            <br clear="all">
            <div><br>
            </div>
            -- <br>
            <div dir="ltr">Sean Dobbs<br>
              Department of Physics & Astronomy <br>
              Northwestern University<br>
              phone: 847-467-2826</div>
          </div>
          _______________________________________________<br>
          Halld-offline mailing list<br>
          <a moz-do-not-send="true" href="mailto:Halld-offline@jlab.org">Halld-offline@jlab.org</a><br>
          <a class="moz-txt-link-freetext" href="https://mailman.jlab.org/mailman/listinfo/halld-offline">https://mailman.jlab.org/mailman/listinfo/halld-offline</a></blockquote>
      </div>
      <br>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
Halld-offline mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Halld-offline@jlab.org">Halld-offline@jlab.org</a>
<a class="moz-txt-link-freetext" href="https://mailman.jlab.org/mailman/listinfo/halld-offline">https://mailman.jlab.org/mailman/listinfo/halld-offline</a></pre>
    </blockquote>
    <br>
  </div>

_______________________________________________<br>Halld-offline mailing list<br><a href="mailto:Halld-offline@jlab.org">Halld-offline@jlab.org</a><br>https://mailman.jlab.org/mailman/listinfo/halld-offline</blockquote></div><br></body></html>