<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">That is great! It's a much more sane patch than my alignment hack. I have verified that is solved the problem on my system too.<div><br></div><div>There are still a lot more failing tests on the 32-bit version dashboard build. It certainly appears to be an optimization bug with this version of GCC. I am not sure how worth wild it will be to pursue the remaining test failures that are occurring here:</div><div><br></div><div><a href="http://open.cdash.org/viewTest.php?onlyfailed&buildid=2355571">http://open.cdash.org/viewTest.php?onlyfailed&buildid=2355571</a></div><div><br></div><div>If we have to just hack the structure of the loops, or move around how the data is laid out in memory.</div><div><br></div><div>The version of this compiler is:</div><div><br></div><div><div>$ gcc-4.7 --version</div><div>gcc-4.7 (Debian 4.7.0-12) 4.7.0</div><div>Copyright (C) 2012 Free Software Foundation, Inc.</div><div>This is free software; see the source for copying conditions. There is NO</div><div>warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.</div><div><br></div></div><div>So it's still very new. Is it likely that Debian is going to release with this version of the compiler? or are the bug is gcc going to get patched?</div><div><br></div><div>Thanks,</div><div>Brad</div><div><br><div><div>On Jun 12, 2012, at 2:44 PM, Bill Lorensen wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div>Paul,<br><br>Can you send me the updated file (not a debian patch) so I can submit<br>it as a gerrit patch?<br><br>Bill<br><br>On Tue, Jun 12, 2012 at 2:28 PM, Paul Novotny <<a href="mailto:paul@paulnovo.us">paul@paulnovo.us</a>> wrote:<br><blockquote type="cite">I spent some time on this too, and again was about to submit a patch<br></blockquote><blockquote type="cite">(man, I am just too slow). The code in ComputeMatrix looked a bit odd to<br></blockquote><blockquote type="cite">me, ie using where the for counter stopped as a condition in the next<br></blockquote><blockquote type="cite">if. I am thinking gcc does some for-loop branching optimization that<br></blockquote><blockquote type="cite">does't like this. Anyway, if you compile with the attached patch, the<br></blockquote><blockquote type="cite">Spatial Object segfaults go away.<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">-Paul<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">On Tue, 2012-06-12 at 13:58 -0400, Bill Lorensen wrote:<br></blockquote><blockquote type="cite"><blockquote type="cite">First, I can't imagine how you figured this out. You are starting to<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">become a legend in your time...<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">Why not #ifdef your solution. Something like:<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">#if (__GNUC__ == 4.7)<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"> InputVectorType m_MatrixScale __attribute__ ((aligned) );<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">#else<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"> InputVectorType m_MatrixScale;<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">#endif<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">On Tue, Jun 12, 2012 at 12:16 PM, Bradley Lowekamp<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><<a href="mailto:blowekamp@mail.nih.gov">blowekamp@mail.nih.gov</a>> wrote:<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Hello Steve,<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">I spent some time looking into this.<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">I downloaded the latest Debian 64-bit and then updated to the sid unstable<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">with gcc 4.7. I think this is the same version you are using?<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">These are the failing tests I had:<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">The following tests FAILED:<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">508 - itkDiffusionTensor3DReconstructionImageFilterTest (SEGFAULT)<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">885 - itkReadWriteSpatialObjectTest (SEGFAULT)<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">886 - itkReadWriteSpatialObjectTest1 (SEGFAULT)<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">887 - itkReadWriteSpatialObjectTest2 (SEGFAULT)<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">1504 - itkMeshSpatialObjectIOTest (SEGFAULT)<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">1505 - itkMeshSpatialObjectIOTest1 (SEGFAULT)<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">1506 - itkMeshSpatialObjectIOTest2 (SEGFAULT)<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">1924 - itkImageSpatialObjectTest (SEGFAULT)<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">1925 - itkSpatialObjectToImageStatisticsCalculatorTest (SEGFAULT)<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">1926 - itkImageMaskSpatialObjectTest (SEGFAULT)<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">1927 - itkImageMaskSpatialObjectTest2 (SEGFAULT)<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">1928 - itkImageMaskSpatialObjectTest3 (SEGFAULT)<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">1931 - itkSceneSpatialObjectTest (SEGFAULT)<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">1933 - itkMetaArrowConverterTest (SEGFAULT)<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">1946 - itkArrowSpatialObjectTest (SEGFAULT)<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">2124 - itkTransformsSetParametersTest (SEGFAULT)<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Errors while running CTest<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">I was able to repeat many of your test failures. I then hacked and debugged<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">the code for a little bit. I don't see anything wrong with it. Based on you<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">experiment of mixing optimization levels and what I observed, I would tend<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">to agree that there is likely some kind of gcc bug, perhaps in intra-unit<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">optimization or how the data get laid out in certain circumstances.<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">The most interesting things I was able to do it the following change:<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">diff --git a/Modules/Core/Transform/include/itkScalableAffineTransform.h<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">b/Modules/Core/Transform/i<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">index 6032127..8dbcdc4 100644<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">--- a/Modules/Core/Transform/include/itkScalableAffineTransform.h<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">+++ b/Modules/Core/Transform/include/itkScalableAffineTransform.h<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">@@ -166,7 +166,7 @@ private:<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"> const Self & operator=(const Self &);<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"> double m_Scale[NDimensions];<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">- InputVectorType m_MatrixScale;<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">+ InputVectorType m_MatrixScale __attribute__ ((aligned) );<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"> }; //class ScalableAffineTransform<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"> } // namespace itk<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">With this change all tests pass. Which is _really_ good to know! This mean<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">that all these failing tests are related to this same bug. So I think I have<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">taken this as far as it needs to... I guess I hope that a patch in gcc 4.7<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">will address this issue.<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">On another note. Have you looking at the SimpleITK project? Have you<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">considered also packaging that for Debian distributions?<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Thanks,<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Brad<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">On Jun 10, 2012, at 1:41 AM, Steve M. Robbins wrote:<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">On Sun, Jun 10, 2012 at 12:03:39AM -0500, Steve M. Robbins wrote:<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Well, I spent some time looking at itkReadWriteSpatialObjectTest and<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">I've determined a few things.<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"> [...]<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">2. It works when built using gcc 4.7 in RelWithDebInfo mode (flags -O2 -g).<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"> [...]<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">4. It fails when built using gcc 4.7 with flags -O3 -g -DNDEBUG.<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">One further item. Of the three files that make up<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">ITKIOSpatialObjectsTestDriver, as long as this one<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">is built with -O2:<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">itkPolygonGroupSpatialObjectXMLFileTest.cxx.o<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">I can build the other two with -O3:<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">ITKIOSpatialObjectsTestDriver.cxx.o<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">itkReadWriteSpatialObjectTest.cxx.o<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">and the test completes successfully. This makes no sense to me<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">because itkPolygonGroupSpatialObjectXMLFileTest code is not even<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">executed when running the test in question ("ctest -R<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">itkReadWriteSpatialObjectTest").<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">-Steve<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><signature.asc>_______________________________________________<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Powered by <a href="http://www.kitware.com">www.kitware.com</a><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Visit other Kitware open-source projects at<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><a href="http://www.kitware.com/opensource/opensource.html">http://www.kitware.com/opensource/opensource.html</a><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Kitware offers ITK Training Courses, for more information visit:<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><a href="http://kitware.com/products/protraining.php">http://kitware.com/products/protraining.php</a><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Please keep messages on-topic and check the ITK FAQ at:<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><a href="http://www.itk.org/Wiki/ITK_FAQ">http://www.itk.org/Wiki/ITK_FAQ</a><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Follow this link to subscribe/unsubscribe:<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><a href="http://www.itk.org/mailman/listinfo/insight-developers">http://www.itk.org/mailman/listinfo/insight-developers</a><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">========================================================<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Bradley Lowekamp<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Medical Science and Computing for<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Office of High Performance Computing and Communications<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">National Library of Medicine<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><a href="mailto:blowekamp@mail.nih.gov">blowekamp@mail.nih.gov</a><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><br></blockquote><br><br><br>-- <br>Unpaid intern in BillsBasement at noware dot com<br></div></blockquote></div><br><div>
<div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; font-size: 12px; "><span class="Apple-style-span" style="border-collapse: separate; -webkit-border-horizontal-spacing: 0px; -webkit-border-vertical-spacing: 0px; color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; -webkit-text-decorations-in-effect: none; text-indent: 0px; -webkit-text-size-adjust: auto; text-transform: none; orphans: 2; white-space: normal; widows: 2; word-spacing: 0px; "><p style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; "><font face="Helvetica" size="3" style="font: normal normal normal 12px/normal Helvetica; ">========================================================</font></p><p style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; "><font face="Helvetica" size="3" style="font: normal normal normal 12px/normal Helvetica; ">Bradley Lowekamp<span class="Apple-converted-space"> </span><span class="Apple-converted-space"> </span></font></p><p style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; "><font face="Helvetica" size="3" style="font: normal normal normal 12px/normal Helvetica; ">Medical Science and Computing for</font></p><p style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; "><font face="Helvetica" size="3" style="font: normal normal normal 12px/normal Helvetica; ">Office of High Performance Computing and Communications</font></p><p style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; "><font face="Helvetica" size="3" style="font: normal normal normal 12px/normal Helvetica; ">National Library of Medicine<span class="Apple-converted-space"> </span></font></p><p style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; "><font face="Helvetica" size="3" style="font: normal normal normal 12px/normal Helvetica; "><a href="mailto:blowekamp@mail.nih.gov">blowekamp@mail.nih.gov</a></font></p><br class="Apple-interchange-newline"></span></div><br class="Apple-interchange-newline">
</div>
<br></div></body></html>