<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    Here it is for the 3 compilers.<br>
    <br>
    GNU FC <a class="moz-txt-link-freetext" href="http://bpaste.net/show/31029/">http://bpaste.net/show/31029/</a><br>
    <br>
    Intel FC <a class="moz-txt-link-freetext" href="http://bpaste.net/show/QJ6k5jc52t7MGUlvrlEV/">http://bpaste.net/show/QJ6k5jc52t7MGUlvrlEV/</a><br>
    <br>
    PGI FC <a class="moz-txt-link-freetext" href="http://bpaste.net/show/DZzY6Re5iXMgLSHydOgJ/">http://bpaste.net/show/DZzY6Re5iXMgLSHydOgJ/</a><br>
    <br>
    <br>
    On 06/08/2012 09:15 AM, John Mousel wrote:
    <blockquote
cite="mid:CA+zegpmB05FbFFzZfQtSk+DkOJETTTUL-crODZfc=svKx=mfLg@mail.gmail.com"
      type="cite">Attaching the Valgrind output would probably help a
      lot.<br>
      <br>
      <div class="gmail_quote">On Fri, Jun 8, 2012 at 9:02 AM, Tabrez
        Ali <span dir="ltr"><<a moz-do-not-send="true"
            href="mailto:stali@geology.wisc.edu" target="_blank">stali@geology.wisc.edu</a>></span>
        wrote:<br>
        <blockquote class="gmail_quote" style="margin:0 0 0
          .8ex;border-left:1px #ccc solid;padding-left:1ex">
          <div bgcolor="#FFFFFF" text="#000000"> Thanks for your answer.
            Unfortunately its not that. <br>
            <br>
            Based on past experience I am sure I am overlooking
            something very simple but I cant seem to find out what. Btw
            METIS 4 worked fine before.<br>
            <br>
            On 06/08/2012 08:39 AM, John Mousel wrote:
            <blockquote type="cite">It's hard to tell from the info you
              provided, but you seem to be playing fast and loose with
              your type declarations. METIS is expecting real_t, which
              is a 32 bit real if you haven't changed the definition in
              metis.h. I know this has caused me problems in the past.<br>
              <br>
              <div class="gmail_quote">On Fri, Jun 8, 2012 at 7:38 AM,
                Tabrez Ali <span dir="ltr"><<a
                    moz-do-not-send="true"
                    href="mailto:stali@geology.wisc.edu" target="_blank">stali@geology.wisc.edu</a>></span>
                wrote:<br>
                <blockquote class="gmail_quote" style="margin:0 0 0
                  .8ex;border-left:1px #ccc solid;padding-left:1ex">
                  Sorry about a question not directly related to PETSc
                  but has anyone here been able to use the METIS 5.0
                  (that PETSc 3.3/dev downloads/builds) with Fortran?
                  There has been an API change from 4 to 5 but I am
                  having some trouble and METIS manual/forums havent
                  been useful.<br>
                  <br>
                  For example consider the simple code (below) that
                  partitions a two element mesh made of linear quads
                  into two. The elements are numbered 0 1 2 3 and 1 4 5
                  2.It works fine with GNU FC (no valgrind errors). With
                  Intel FC it works fines (though valgrind throws a
                  bunch of errors). However with PGI compilers I get a
                  segfault.<br>
                  <br>
                   program test<br>
                   implicit none<br>
                   integer, parameter   :: nels=2, nnds=6, npel=4<br>
                   integer              :: eptr(nels+1),
                  nodes(nels*npel), epart(nels), npart(nnds), n<br>
                   integer, pointer     :: vwgt(:)=>null(),
                  vsize(:)=>null(), mopts(:)=>null()<br>
                   real(8), pointer     :: tpwgts(:)=>null()<br>
                   eptr=(/0,4,7/)<br>
                   nodes=(/0,1,2,3,1,4,5,2/)<br>
                   call
METIS_PartMeshNodal(nels,nnds,eptr,nodes,vwgt,vsize,2,tpwgts,mopts,n,epart,npart)<br>
                   print*, npart; print*, epart<br>
                  end program test<br>
                  <br>
                  According to the manual moving from METIS 4 to 5 only
                  involves passing some additional nulls. I am not sure
                  what I missed.<br>
                  <br>
                  <a moz-do-not-send="true"
                    href="http://glaros.dtc.umn.edu/gkhome/fetch/sw/metis/manual.pdf"
                    target="_blank">http://glaros.dtc.umn.edu/gkhome/fetch/sw/metis/manual.pdf</a><br>
                  <br>
                  Thanks in advance.<span><font color="#888888"><br>
                      <br>
                      Tabrez<span class="HOEnZb"><font color="#888888"><br>
                        </font></span></font></span></blockquote>
                <span class="HOEnZb"><font color="#888888"> </font></span></div>
              <span class="HOEnZb"><font color="#888888"> </font></span></blockquote>
            <span class="HOEnZb"><font color="#888888"> <br>
                <pre cols="72">-- 
No one trusts a model except the one who wrote it; Everyone trusts an observation except the one who made it- Harlow Shapley</pre>
              </font></span></div>
        </blockquote>
      </div>
      <br>
    </blockquote>
    <br>
  </body>
</html>