<html>
  <head>
    <meta content="text/html; charset=utf-8" http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <p>Thank you for your reply. The source code I have used is from the
      website of PETSc, not from the git repository.</p>
    <p>I will have a test with the code from git repository.<br>
    </p>
    <p><br>
    </p>
    <p>leejearl<br>
    </p>
    <br>
    <div class="moz-cite-prefix">On 2016年08月13日 08:49, Oxberry, Geoffrey
      Malcolm wrote:<br>
    </div>
    <blockquote cite="mid:F6F1177B-F91D-4089-A112-86B0FC1834D7@llnl.gov"
      type="cite">
      <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
      <br class="">
      <div>
        <blockquote type="cite" class="">
          <div class="">On Aug 12, 2016, at 5:41 PM, leejearl <<a
              moz-do-not-send="true" href="mailto:leejearl@126.com"
              class="">leejearl@126.com</a>> wrote:</div>
          <br class="Apple-interchange-newline">
          <div class="">
            <div bgcolor="#FFFFFF" text="#000000" class="">
              <p class="">Hi, Matt:<br class="">
              </p>
                  <br class="">
              <div class=""><br class="">
              </div>
              > Can you verify that you are running the master
              branch?<br class="">
            </div>
          </div>
        </blockquote>
        <div><br class="">
        </div>
        <div>cd ${PETSC_DIR}</div>
        <div>git branch</div>
        <div><br class="">
        </div>
        <div>The last command should return something like a list of
          branch names, and the branch name with an asterisk to the left
          of it will be the branch you are currently on.</div>
        <div><br class="">
        </div>
        <div>Geoff</div>
        <div>
          <div><br class="">
          </div>
        </div>
        <blockquote type="cite" class="">
          <div class="">
            <div bgcolor="#FFFFFF" text="#000000" class="">I am not
              sure, how can I verify this?<br class="">
              And I configure PETSc with this command <br class="">
              "./configure --prefix=$HOME/Install/petsc-openmpi
              --with-mpi=/home/leejearl/Install/openmpi/gnu/1.8.4
              --download-exodusii=yes --download-netcdf
              --with-hdf5-dir=/home/leejearl/Install/hdf5-1.8.14
              --download-metis=yes".<br class="">
              Is there some problem? Can you show me your command for
              configuring PETSc?<br class="">
              <br class="">
              <br class="">
              Thanks<br class="">
              <br class="">
              leejearl<br class="">
              <br class="">
              <br class="">
              <br class="">
              <br class="">
              <br class="">
              <div class="moz-cite-prefix">On 2016年08月13日 01:10, Matthew
                Knepley wrote:<br class="">
              </div>
              <blockquote
cite="mid:CAMYG4GmtHnQiAmWcZ9Pdb2ZMg9M2Pj+14kqnjd5iM4KmzyuRhw@mail.gmail.com"
                type="cite" class="">
                <div dir="ltr" class="">
                  <div class="gmail_extra">
                    <div class="gmail_quote">On Thu, Aug 11, 2016 at
                      8:00 PM, leejearl <span dir="ltr" class="">
                        <<a moz-do-not-send="true"
                          href="mailto:leejearl@126.com" target="_blank"
                          class="">leejearl@126.com</a>></span>
                      wrote:<br class="">
                      <blockquote class="gmail_quote" style="margin:0px
                        0px 0px 0.8ex;border-left:1px solid
                        rgb(204,204,204);padding-left:1ex">
                        <div bgcolor="#FFFFFF" text="#000000" class="">
                          <p class="">Thank you for your reply. I have
                            attached the code, grid and the error
                            message.</p>
                          <p class="">cavity.c is the code file,
                            cavity.exo is the grid, and error.dat is the
                            error message.</p>
                          <p class="">The command is "mpirun -n 2
                            ./cavity</p>
                        </div>
                      </blockquote>
                      <div class=""><br class="">
                      </div>
                      <div class="">Can you verify that you are running
                        the master branch? I just ran this and got</div>
                      <div class=""><br class="">
                      </div>
                      <div class="">
                        <div class="">DM Object: 2 MPI processes</div>
                        <div class="">  type: plex</div>
                        <div class="">DM_0x84000004_0 in 2 dimensions:</div>
                        <div class="">  0-cells: 5253 5252</div>
                        <div class="">  1-cells: 10352 10350</div>
                        <div class="">  2-cells: 5298 (198) 5297 (198)</div>
                        <div class="">Labels:</div>
                        <div class="">  ghost: 2 strata of sizes (199,
                          400)</div>
                        <div class="">  vtk: 1 strata of sizes (4901)</div>
                        <div class="">  Cell Sets: 1 strata of sizes
                          (5100)</div>
                        <div class="">  Face Sets: 3 strata of sizes
                          (53, 99, 50)</div>
                        <div class="">  depth: 3 strata of sizes (5253,
                          10352, 5298)</div>
                      </div>
                      <div class=""><br class="">
                      </div>
                      <div class="">  Thanks,</div>
                      <div class=""><br class="">
                      </div>
                      <div class="">     Matt</div>
                      <div class=""> </div>
                      <blockquote class="gmail_quote" style="margin:0px
                        0px 0px 0.8ex;border-left:1px solid
                        rgb(204,204,204);padding-left:1ex">
                        <div bgcolor="#FFFFFF" text="#000000" class="">
                          <div class="">On 2016年08月11日 23:29, Matthew
                            Knepley wrote:<br class="">
                          </div>
                          <blockquote type="cite" class="">
                            <div dir="ltr" class="">
                              <div class="gmail_extra">
                                <div class="gmail_quote">On Thu, Aug 11,
                                  2016 at 3:14 AM, leejearl <span
                                    dir="ltr" class="">
                                    <<a moz-do-not-send="true"
                                      href="mailto:leejearl@126.com"
                                      target="_blank" class="">leejearl@126.com</a>></span>
                                  wrote:<br class="">
                                  <blockquote class="gmail_quote"
                                    style="margin:0px 0px 0px
                                    0.8ex;border-left:1px solid
                                    rgb(204,204,204);padding-left:1ex">
                                    <div bgcolor="#FFFFFF"
                                      text="#000000" class="">
                                      <p class="">Hi, <br class="">
                                            Thank you for your reply. It
                                        help me very much.<br class="">
                                            But, for
                                        "/petsc-3.7.2/src/ts/examples/<wbr
                                          class="">tutorials/ex11.c",
                                        when I set the overlap to 2
                                        levels, the command is
                                        <br class="">
                                        "mpirun -n 3 ./ex11 -f
                                        annulus-20.exo -ufv_mesh_overlap
                                        2 -physics sw", it suffers a
                                        error.<br class="">
                                            It seems to me that setting
                                        overlap to 2 is very common. Are
                                        there issues that I have not
                                        take into consideration?<br
                                          class="">
                                            Any help are appreciated.</p>
                                    </div>
                                  </blockquote>
                                  <div class="">I will check this out. I
                                    have not tested an overlap of 2 here
                                    since I generally use nearest
                                    neighbor FV methods for</div>
                                  <div class="">unstructured stuff. I
                                    have test examples that run fine for
                                    overlap > 1. Can you send the
                                    entire error message?</div>
                                  <div class=""><br class="">
                                  </div>
                                  <div class="">If the error is not in
                                    the distribution, but rather in the
                                    analytics, that is understandable
                                    because this example is only</div>
                                  <div class="">intended to be run using
                                    a nearest neighbor FV method, and
                                    thus might be confused if we give it
                                    two layers of ghost</div>
                                  <div class="">cells.</div>
                                  <div class=""><br class="">
                                  </div>
                                  <div class="">   Matt</div>
                                  <div class=""> </div>
                                  <blockquote class="gmail_quote"
                                    style="margin:0px 0px 0px
                                    0.8ex;border-left:1px solid
                                    rgb(204,204,204);padding-left:1ex">
                                    <div bgcolor="#FFFFFF"
                                      text="#000000" class="">
                                      <p class=""><span class=""><font
                                            class="" color="#888888"><br
                                              class="">
                                            leejearl</font></span></p>
                                      <div class="">
                                        <div class=""><br class="">
                                          <div class="">On 2016年08月11日
                                            14:57, Julian Andrej wrote:<br
                                              class="">
                                          </div>
                                          <blockquote type="cite"
                                            class="">
                                            <div dir="ltr" class="">Hi,
                                              <div class=""><br class="">
                                              </div>
                                              <div class="">take a look
                                                at slide 10 of [1],
                                                there is visually
                                                explained what the
                                                overlap between
                                                partitions is.</div>
                                              <div class=""><br class="">
                                              </div>
                                              <div class="">[1] <a
                                                  moz-do-not-send="true"
href="https://www.archer.ac.uk/training/virtual/files/2015/06-PETSc/slides.pdf"
                                                  target="_blank"
                                                  class="">https://www.archer.ac.uk/t<wbr
                                                    class="">raining/virtual/files/2015/06-<wbr
                                                    class="">PETSc/slides.pdf</a></div>
                                            </div>
                                            <div class="gmail_extra"><br
                                                class="">
                                              <div class="gmail_quote">On
                                                Thu, Aug 11, 2016 at
                                                8:48 AM, leejearl <span
                                                  dir="ltr" class="">
                                                  <<a
                                                    moz-do-not-send="true"
href="mailto:leejearl@126.com" target="_blank" class="">leejearl@126.com</a>></span>
                                                wrote:<br class="">
                                                <blockquote
                                                  class="gmail_quote"
                                                  style="margin:0px 0px
                                                  0px
                                                  0.8ex;border-left:1px
                                                  solid
                                                  rgb(204,204,204);padding-left:1ex">
                                                  Hi, all:<br class="">
                                                      I want to use
                                                  PETSc to build my FVM
                                                  code. Now, I have a
                                                  question about<br
                                                    class="">
                                                  the function 
                                                  DMPlexDistribute(DM
                                                  dm, PetscInt overlap,
                                                  PetscSF *sf, DM
                                                  *dmOverlap) .<br
                                                    class="">
                                                  <br class="">
                                                      In the example
                                                  "/petsc-3.7.2/src/ts/examples/<wbr
                                                    class="">tutorials/ex11.c".
                                                  When I set the overlap<br
                                                    class="">
                                                  as 0 or 1, it works
                                                  well. But, if I set
                                                  the overlap as 2, it
                                                  suffers a problem.<br
                                                    class="">
                                                      I am confused
                                                  about the value of
                                                  overlap. Can it be set
                                                  as 2? What is the
                                                  meaning of<br class="">
                                                  the parameter overlap?<br
                                                    class="">
                                                      Any helps are
                                                  appreciated!<span
                                                    class=""><font
                                                      class=""
                                                      color="#888888"><br
                                                        class="">
                                                      <br class="">
                                                      leejearl<br
                                                        class="">
                                                      <br class="">
                                                      <br class="">
                                                      <br class="">
                                                    </font></span></blockquote>
                                              </div>
                                              <br class="">
                                            </div>
                                          </blockquote>
                                          <br class="">
                                        </div>
                                      </div>
                                    </div>
                                  </blockquote>
                                </div>
                                <br class="">
                                <br class="" clear="all">
                                <span class=""><font class=""
                                    color="#888888">
                                    <div class=""><br class="">
                                    </div>
                                    -- <br class="">
                                    <div
                                      data-smartmail="gmail_signature"
                                      class="">What most experimenters
                                      take for granted before they begin
                                      their experiments is infinitely
                                      more interesting than any results
                                      to which their experiments lead.<br
                                        class="">
                                      -- Norbert Wiener</div>
                                  </font></span></div>
                            </div>
                          </blockquote>
                          <br class="">
                        </div>
                      </blockquote>
                    </div>
                    <br class="">
                    <br class="" clear="all">
                    <div class=""><br class="">
                    </div>
                    -- <br class="">
                    <div class="gmail_signature"
                      data-smartmail="gmail_signature">What most
                      experimenters take for granted before they begin
                      their experiments is infinitely more interesting
                      than any results to which their experiments lead.<br
                        class="">
                      -- Norbert Wiener</div>
                  </div>
                </div>
              </blockquote>
              <br class="">
            </div>
          </div>
        </blockquote>
      </div>
      <br class="">
    </blockquote>
    <br>
    <pre class="moz-signature" cols="72">-- 
李季
西北工业大学航空学院流体力学系
Phone: 13324530085
QQ: 188524324
</pre>
  </body>
</html>