<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    On 2/23/2012 7:24 PM, Matthew Knepley wrote:
    <blockquote
cite="mid:CAMYG4G=x=6vrEh6EhAe9JSkHdoiwFdtJOZeUcxDTRYutQMN2ug@mail.gmail.com"
      type="cite">On Thu, Feb 23, 2012 at 12:05 PM, Bojan Niceno <span
        dir="ltr">&lt;<a moz-do-not-send="true"
          href="mailto:bojan.niceno@psi.ch">bojan.niceno@psi.ch</a>&gt;</span>
      wrote:<br>
      <div class="gmail_quote">
        <blockquote class="gmail_quote" style="margin:0 0 0
          .8ex;border-left:1px #ccc solid;padding-left:1ex">
          <div bgcolor="#FFFFFF" text="#000000"> Dear Matthew,<br>
            <br>
            <br>
            thank you for your response.&nbsp; When I use VecCreateGhost, I
            get the following:<br>
          </div>
        </blockquote>
        <div><br>
        </div>
        <div>It appears that you passed a bad communicator. Did you not
          initialize a 'comm' variable?</div>
      </div>
    </blockquote>
    <br>
    I pass PETSC_COMM_WORLD to VecCreateGhost.&nbsp; <br>
    <br>
    I don't know what you mean by 'comm' variable :-(&nbsp; I called all the
    routines to initialize PETSc.<br>
    <br>
    &nbsp;&nbsp;&nbsp; <br>
    &nbsp;&nbsp;&nbsp; Cheers,<br>
    <br>
    &nbsp;&nbsp;&nbsp; <br>
    &nbsp;&nbsp;&nbsp; Bojan<br>
    <br>
    <blockquote
cite="mid:CAMYG4G=x=6vrEh6EhAe9JSkHdoiwFdtJOZeUcxDTRYutQMN2ug@mail.gmail.com"
      type="cite">
      <div class="gmail_quote">
        <div><br>
        </div>
        <div>&nbsp; &nbsp;Matt</div>
        <div>&nbsp;</div>
        <blockquote class="gmail_quote" style="margin:0 0 0
          .8ex;border-left:1px #ccc solid;padding-left:1ex">
          <div bgcolor="#FFFFFF" text="#000000"> [0]PETSC ERROR:
            ------------------------------------------------------------------------<br>
            [0]PETSC ERROR: Caught signal number 11 SEGV: Segmentation
            Violation, probably memory access out of range<br>
            [0]PETSC ERROR: Try option -start_in_debugger or
            -on_error_attach_debugger<br>
            [0]PETSC ERROR: or see <a moz-do-not-send="true"
href="http://www.mcs.anl.gov/petsc/petsc-as/documentation/faq.html#valgrind"
              target="_blank">http://www.mcs.anl.gov/petsc/petsc-as/documentation/faq.html#valgrind</a>[0]PETSC

            ERROR: or try <a moz-do-not-send="true"
              href="http://valgrind.org" target="_blank">http://valgrind.org</a>
            on GNU/linux and Apple Mac OS X to find memory corruption
            errors<br>
            [0]PETSC ERROR: likely location of problem given in stack
            below<br>
            [0]PETSC ERROR: ---------------------&nbsp; Stack Frames
            ------------------------------------<br>
            [0]PETSC ERROR: Note: The EXACT line numbers in the stack
            are not available,<br>
            [0]PETSC ERROR:&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; INSTEAD the line number of the start
            of the function<br>
            [0]PETSC ERROR:&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; is given.<br>
            [0]PETSC ERROR: [0] PetscCommDuplicate line 140
            src/sys/objects/tagm.c<br>
            [0]PETSC ERROR: [0] PetscHeaderCreate_Private line 30
            src/sys/objects/inherit.c<br>
            [0]PETSC ERROR: [0] VecCreate line 32
            src/vec/vec/interface/veccreate.c<br>
            [0]PETSC ERROR: [0] VecCreateGhostWithArray line 567
            src/vec/vec/impls/mpi/pbvec.c<br>
            [0]PETSC ERROR: [0] VecCreateGhost line 647
            src/vec/vec/impls/mpi/pbvec.c<br>
            [0]PETSC ERROR: --------------------- Error Message
            ------------------------------------<br>
            [0]PETSC ERROR: Signal received!<br>
            [0]PETSC ERROR:
            ------------------------------------------------------------------------<br>
            [0]PETSC ERROR: Petsc Release Version 3.2.0, Patch 6, Wed
            Jan 11 09:28:45 CST 2012<br>
            [0]PETSC ERROR: See docs/changes/index.html for recent
            updates.<br>
            [0]PETSC ERROR: See docs/faq.html for hints about trouble
            shooting.<br>
            [0]PETSC ERROR: See docs/index.html for manual pages.<br>
            [0]PETSC ERROR:
            ------------------------------------------------------------------------<br>
            [0]PETSC ERROR: ./PSI-Flow on a arch-linu named lccfd06 by
            niceno Thu Feb 23 19:02:45 2012<br>
            [0]PETSC ERROR: Libraries linked from
            /homecfd/niceno/PETSc-3.2-p6/arch-linux2-c-debug/lib<br>
            [0]PETSC ERROR: Configure run at Fri Feb 10 10:24:13 2012<br>
            [0]PETSC ERROR: Configure options<br>
            [0]PETSC ERROR:
            ------------------------------------------------------------------------<br>
            [0]PETSC ERROR: User provided function() line 0 in unknown
            directory unknown file<br>
            <br>
            I don't understand what could be causing it.&nbsp; I took very
            good care to match the global numbers of ghost cells when
            calling VecCreateGhost<br>
            <br>
            <br>
            &nbsp;&nbsp;&nbsp; Kind regards,<br>
            <br>
            <br>
            &nbsp;&nbsp;&nbsp; Bojan<br>
            <br>
            <br>
            On 2/23/2012 5:53 PM, Matthew Knepley wrote:
            <blockquote type="cite">On Thu, Feb 23, 2012 at 10:46 AM,
              Bojan Niceno <span dir="ltr">&lt;<a
                  moz-do-not-send="true"
                  href="mailto:bojan.niceno@psi.ch" target="_blank">bojan.niceno@psi.ch</a>&gt;</span>
              wrote:<br>
              <div class="gmail_quote">
                <blockquote class="gmail_quote" style="margin:0 0 0
                  .8ex;border-left:1px #ccc solid;padding-left:1ex"> Hi
                  all,<br>
                  <br>
                  I've never used a mailing list before, so I hope this
                  message will reach PETSc users and experts and someone
                  might be willing to help me. &nbsp;I am also novice in
                  PETSc.<br>
                  <br>
                  I have developed an unstructured finite volume solver
                  on top of PETSc libraries. &nbsp;In sequential, it works
                  like a charm. &nbsp;For the parallel version, I do domain
                  decomposition externally with Metis, and work out
                  local and global numberings, as well as communication
                  patterns between processor. &nbsp;(The latter don't seem to
                  be needed for PETSc, though.) &nbsp;When I run my program
                  in parallel, it also works, but I miss values in
                  vectors' ghost points.<br>
                  <br>
                  I create vectors with command:
                  VecCreate(PETSC_COMM_WORLD, &amp;x);<br>
                  <br>
                  Is it possible to get the ghost values if a vector is
                  created like this?<br>
                </blockquote>
                <div><br>
                </div>
                <div>I do not understand this question. By definition,
                  "ghost values" are those not stored in the global
                  vector.</div>
                <div>&nbsp;</div>
                <blockquote class="gmail_quote" style="margin:0 0 0
                  .8ex;border-left:1px #ccc solid;padding-left:1ex"> I
                  have tried to use VecCreateGhost, but for some reason
                  which is beyond my comprehension, PETSc goes berserk
                  when it reaches the command:
                  VecCreateGhost(PETSC_COMM_WORLD, n, PETSC_DECIDE,
                  nghost, ifrom, &amp;x)<br>
                </blockquote>
                <div><br>
                </div>
                <div>I think you can understand that "berserk" tells me
                  absolutely nothing. Error message? Stack trace? Did
                  you try to run an</div>
                <div>example which uses VecGhost?</div>
                <div><br>
                </div>
                <div>&nbsp; Thanks,</div>
                <div><br>
                </div>
                <div>&nbsp; &nbsp; &nbsp;Matt</div>
                <div>&nbsp;</div>
                <blockquote class="gmail_quote" style="margin:0 0 0
                  .8ex;border-left:1px #ccc solid;padding-left:1ex"> Can
                  anyone help me? &nbsp;Either how to reach ghost values for
                  vector created by VecCreate, or how to use
                  VecCreateGhost properly?<br>
                  <br>
                  <br>
                  &nbsp; Kind regards,<br>
                  <br>
                  &nbsp; Bojan<br>
                </blockquote>
              </div>
              <br>
              <br clear="all">
              <span class="HOEnZb"><font color="#888888">
                  <div><br>
                  </div>
                  -- <br>
                  What most experimenters take for granted before they
                  begin their experiments is infinitely more interesting
                  than any results to which their experiments lead.<br>
                  -- Norbert Wiener<br>
                </font></span></blockquote>
            <span class="HOEnZb"><font color="#888888"> <br>
                <br>
                <div>-- <br>
                  <img src="cid:part1.05030409.07030809@psi.ch"
                    border="0"></div>
              </font></span></div>
        </blockquote>
      </div>
      <br>
      <br clear="all">
      <div><br>
      </div>
      -- <br>
      What most experimenters take for granted before they begin their
      experiments is infinitely more interesting than any results to
      which their experiments lead.<br>
      -- Norbert Wiener<br>
    </blockquote>
    <br>
    <br>
    <div class="moz-signature">-- <br>
      <img src="cid:part2.06020705.05020108@psi.ch" border="0"></div>
  </body>
</html>