<html><head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    Hi Eric,<br>
    <br>
    1 hour is normal for running the entire MPICH test suite. It looks
    like your failures are either dynamic process (e.g., spawn) or fault
    tolerance related tests. They should be irrelevant to the PETSc
    issue.<br>
    <br>
    Min<br>
    <div class="moz-cite-prefix">On 2018/03/27 21:09, Eric Chamberland
      wrote:<br>
    </div>
    <blockquote type="cite" cite="mid:7c2ecc9c-184f-5ae6-18bb-bf595a7df19f@giref.ulaval.ca">
      
      <p>Hi Min and Matthew,</p>
      <p>In fact, I just ran an "hello word" on 2 processes and it
        works.  I do not have a more complicated example without PETSc
        since I have a Petsc-based source code...</p>
      <p>However, I just tried to launch "make testing" into the mpich
        directory and it ended with some failed tests  and it was very
        long: about an hour.  Is it normal?<br>
      </p>
      <p>Please, see the attached file: summary.xml</p>
      <p>Thanks,</p>
      <p>Eric</p>
      <p><br>
      </p>
      <br>
      <p><br>
      </p>
      <br>
      <div class="moz-cite-prefix">On 27/03/18 05:04 PM, Matthew Knepley
        wrote:<br>
      </div>
      <blockquote type="cite" cite="mid:CAMYG4G=_U5HVuPKNCyX2Vu+EpuLUWaRgQFKsNabpzZV2jJ97Zw@mail.gmail.com">
        <div dir="ltr">
          <div class="gmail_extra">
            <div class="gmail_quote">On Tue, Mar 27, 2018 at 4:59 PM,
              Min Si <span dir="ltr"><<a href="mailto:msi@anl.gov" target="_blank" moz-do-not-send="true">msi@anl.gov</a>></span>
              wrote:<br>
              <blockquote class="gmail_quote" style="margin:0 0 0
                .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi
                Eric,<br>
                <br>
                It will be great if you could give us a simple MPI
                program (not with PETSc) to reproduce this issue. If
                this is a problem happens only when PETSc is involved,
                the PETSc team can give you more suggestions.<br>
              </blockquote>
              <div><br>
              </div>
              <div>Hi Min,</div>
              <div><br>
              </div>
              <div>It is really easy to run PETSc at ANL. I am sure one
                of us can help if you cannot reproduce this bug on your
                own.</div>
              <div><br>
              </div>
              <div>  Thanks,</div>
              <div><br>
              </div>
              <div>     Matt</div>
              <div> </div>
              <blockquote class="gmail_quote" style="margin:0 0 0
                .8ex;border-left:1px #ccc solid;padding-left:1ex">
                Thanks,<br>
                Min
                <div>
                  <div class="h5"><br>
                    <br>
                    On 2018/03/27 15:38, Eric Chamberland wrote:<br>
                  </div>
                </div>
                <blockquote class="gmail_quote" style="margin:0 0 0
                  .8ex;border-left:1px #ccc solid;padding-left:1ex">
                  <div>
                    <div class="h5"> Hi,<br>
                      <br>
                      since more than 2 weeks that the master branch of
                      mpich is still and it can be reproduced with a
                      simple "make test" after a fresh installation of
                      PETSc...<br>
                      <br>
                      Is anyone testing it?<br>
                      <br>
                      Is it supposed to be working?<br>
                      <br>
                      Just tell me if I should "follow" another mpich
                      branch please.<br>
                      <br>
                      Thanks,<br>
                      <br>
                      Eric<br>
                      <br>
                      <br>
                      On 14/03/18 03:35 AM, Eric Chamberland wrote:<br>
                      <blockquote class="gmail_quote" style="margin:0 0
                        0 .8ex;border-left:1px #ccc
                        solid;padding-left:1ex"> Hi,<br>
                        <br>
                        fwiw, the actual mpich/master branch doesn't
                        passes the PETSc "make test" after a fresh
                        installation...  It hangs just afer the 1 MPI
                        process test, meaning it is locked into the 2
                        process test:<br>
                        <br>
                        make PETSC_DIR=/pmi/cmpbib/compilat<wbr>ion_BIB_dernier_mpich/COMPILE_<wbr>AUTO/mpich-3.x-debug/petsc-3.<wbr>8.3-debug
                        PETSC_ARCH=arch-linux2-c-debug test<br>
                        Running test examples to verify correct
                        installation<br>
                        Using PETSC_DIR=/pmi/cmpbib/compilat<wbr>ion_BIB_dernier_mpich/COMPILE_<wbr>AUTO/mpich-3.x-debug/petsc-3.<wbr>8.3-debug
                        and PETSC_ARCH=arch-linux2-c-debug<br>
                        C/C++ example src/snes/examples/tutorials/ex<wbr>19
                        run successfully with 1 MPI process<br>
                        <br>
                        <br>
                        <br>
                        <br>
                        ^Cmakefile:151: recipe for target 'test' failed<br>
                        make: [test] Interrupt (ignored)<br>
                        <br>
                        thanks,<br>
                        <br>
                        Eric<br>
                        <br>
                        On 13/03/18 08:07 AM, Eric Chamberland wrote:<br>
                        <blockquote class="gmail_quote" style="margin:0
                          0 0 .8ex;border-left:1px #ccc
                          solid;padding-left:1ex"> <br>
                          Hi,<br>
                          <br>
                          each night we are testing mpich/master with
                          our petsc-based code. I don't know if PETSc
                          team is doing the same thing with
                          mpich/master?   (Maybe it is a good idea?)<br>
                          <br>
                          Everything was fine (except the issue <a href="https://github.com/pmodels/mpich/issues/2892" rel="noreferrer" target="_blank" moz-do-not-send="true">https://github.com/pmodels/mpi<wbr>ch/issues/2892</a>)
                          up to commit 7b8d64debd, but since commit
                          mpich:a8a2b30fd21), I have a segfault on a any
                          parallel nightly test.<br>
                          <br>
                          For example, a 2 process test ends at almost
                          different execution points:<br>
                          <br>
                          rank 0:<br>
                          <br>
                          #003: /lib64/libpthread.so.0(+0xf870<wbr>)
                          [0x7f25bf908870]<br>
                          #004: /pmi/cmpbib/compilation_BIB_de<wbr>rnier_mpich/COMPILE_AUTO/BIB/<wbr>bin/BIBMEFGD.opt()
                          [0x64a788]<br>
                          #005: /lib64/libc.so.6(+0x35140)
                          [0x7f25bca18140]<br>
                          #006: /lib64/libc.so.6(__poll+0x2d)
                          [0x7f25bcabfbfd]<br>
                          #007: /opt/mpich-3.x_debug/lib/libmp<wbr>i.so.0(+0x1e4cc9)
                          [0x7f25bd90ccc9]<br>
                          #008: /opt/mpich-3.x_debug/lib/libmp<wbr>i.so.0(+0x1ea55c)
                          [0x7f25bd91255c]<br>
                          #009: /opt/mpich-3.x_debug/lib/libmp<wbr>i.so.0(+0xba657)
                          [0x7f25bd7e2657]<br>
                          #010: /opt/mpich-3.x_debug/lib/libmp<wbr>i.so.0(PMPI_Waitall+0xe3)
                          [0x7f25bd7e3343]<br>
                          #011: /opt/petsc-3.8.3_debug_mpich-3<wbr>.x_debug/lib/libpetsc.so.3.8(P<wbr>etscGatherMessageLengths+0x654<wbr>)
                          [0x7f25c4bb3193]<br>
                          #012: /opt/petsc-3.8.3_debug_mpich-3<wbr>.x_debug/lib/libpetsc.so.3.8(V<wbr>ecScatterCreate_PtoS+0x859)
                          [0x7f25c4e82d7f]<br>
                          #013: /opt/petsc-3.8.3_debug_mpich-3<wbr>.x_debug/lib/libpetsc.so.3.8(V<wbr>ecScatterCreate+0x5684)
                          [0x7f25c4e4d055]<br>
                          #014: /opt/petsc-3.8.3_debug_mpich-3<wbr>.x_debug/lib/libpetsc.so.3.8(V<wbr>ecCreateGhostWithArray+0x688)
                          [0x7f25c4e01a39]<br>
                          #015: /opt/petsc-3.8.3_debug_mpich-3<wbr>.x_debug/lib/libpetsc.so.3.8(V<wbr>ecCreateGhost+0x179)
                          [0x7f25c4e020f6]<br>
                          <br>
                          rank 1:<br>
                          <br>
                          #002: /pmi/cmpbib/compilation_BIB_de<wbr>rnier_mpich/COMPILE_AUTO/GIREF<wbr>/lib/libgiref_opt_Util.so(<wbr>traitementSignal+0x2bd0)
                          [0x7f62df8e7310]<br>
                          #003: /lib64/libc.so.6(+0x35140)
                          [0x7f62d3bc9140]<br>
                          #004: /lib64/libc.so.6(__poll+0x2d)
                          [0x7f62d3c70bfd]<br>
                          #005: /opt/mpich-3.x_debug/lib/libmp<wbr>i.so.0(+0x1e4cc9)
                          [0x7f62d4abdcc9]<br>
                          #006: /opt/mpich-3.x_debug/lib/libmp<wbr>i.so.0(+0x1ea55c)
                          [0x7f62d4ac355c]<br>
                          #007: /opt/mpich-3.x_debug/lib/libmp<wbr>i.so.0(+0x12c9c5)
                          [0x7f62d4a059c5]<br>
                          #008: /opt/mpich-3.x_debug/lib/libmp<wbr>i.so.0(+0x12e102)
                          [0x7f62d4a07102]<br>
                          #009: /opt/mpich-3.x_debug/lib/libmp<wbr>i.so.0(+0xf17a1)
                          [0x7f62d49ca7a1]<br>
                          #010: /opt/mpich-3.x_debug/lib/libmp<wbr>i.so.0(+0x3facf)
                          [0x7f62d4918acf]<br>
                          #011: /opt/mpich-3.x_debug/lib/libmp<wbr>i.so.0(+0x3fc3d)
                          [0x7f62d4918c3d]<br>
                          #012: /opt/mpich-3.x_debug/lib/libmp<wbr>i.so.0(+0xf18d8)
                          [0x7f62d49ca8d8]<br>
                          #013: /opt/mpich-3.x_debug/lib/libmp<wbr>i.so.0(+0x3fb88)
                          [0x7f62d4918b88]<br>
                          #014: /opt/mpich-3.x_debug/lib/libmp<wbr>i.so.0(+0x3fc3d)
                          [0x7f62d4918c3d]<br>
                          #015: /opt/mpich-3.x_debug/lib/libmp<wbr>i.so.0(MPI_Barrier+0x27b)
                          [0x7f62d4918edb]<br>
                          #016: /opt/petsc-3.8.3_debug_mpich-3<wbr>.x_debug/lib/libpetsc.so.3.8(P<wbr>etscCommGetNewTag+0x3ff)
                          [0x7f62dbceb055]<br>
                          #017: /opt/petsc-3.8.3_debug_mpich-3<wbr>.x_debug/lib/libpetsc.so.3.8(P<wbr>etscObjectGetNewTag+0x15d)
                          [0x7f62dbceaadb]<br>
                          #018: /opt/petsc-3.8.3_debug_mpich-3<wbr>.x_debug/lib/libpetsc.so.3.8(V<wbr>ecScatterCreateCommon_PtoS+0x1<wbr>ee)
                          [0x7f62dc03625c]<br>
                          #019: /opt/petsc-3.8.3_debug_mpich-3<wbr>.x_debug/lib/libpetsc.so.3.8(V<wbr>ecScatterCreate_PtoS+0x29c4)
                          [0x7f62dc035eea]<br>
                          #020: /opt/petsc-3.8.3_debug_mpich-3<wbr>.x_debug/lib/libpetsc.so.3.8(V<wbr>ecScatterCreate+0x5684)
                          [0x7f62dbffe055]<br>
                          #021: /opt/petsc-3.8.3_debug_mpich-3<wbr>.x_debug/lib/libpetsc.so.3.8(V<wbr>ecCreateGhostWithArray+0x688)
                          [0x7f62dbfb2a39]<br>
                          #022: /opt/petsc-3.8.3_debug_mpich-3<wbr>.x_debug/lib/libpetsc.so.3.8(V<wbr>ecCreateGhost+0x179)
                          [0x7f62dbfb30f6]<br>
                          <br>
                          Have some other users (PETSc users?) reported
                          problem?<br>
                          <br>
                          Thanks,<br>
                          <br>
                          Eric<br>
                          <br>
                          ps: usual informations:<br>
                          <br>
                          mpich logs:<br>
                          <a href="http://www.giref.ulaval.ca/%7Ecmpgiref/dernier_mpich/2018.03.12.05h39m54s_config.log" rel="noreferrer" target="_blank" moz-do-not-send="true">http://www.giref.ulaval.ca/~cm<wbr>pgiref/dernier_mpich/2018.03.<wbr>12.05h39m54s_config.log</a>
                          <br>
                          <a href="http://www.giref.ulaval.ca/%7Ecmpgiref/dernier_mpich/2018.03.12.05h39m54s_config.system" rel="noreferrer" target="_blank" moz-do-not-send="true">http://www.giref.ulaval.ca/~cm<wbr>pgiref/dernier_mpich/2018.03.<wbr>12.05h39m54s_config.system</a>
                          <br>
                          <a href="http://www.giref.ulaval.ca/%7Ecmpgiref/dernier_mpich/2018.03.12.05h39m54s_mpich_version.txt" rel="noreferrer" target="_blank" moz-do-not-send="true">http://www.giref.ulaval.ca/~cm<wbr>pgiref/dernier_mpich/2018.03.<wbr>12.05h39m54s_mpich_version.txt</a>
                          <br>
                          <a href="http://www.giref.ulaval.ca/%7Ecmpgiref/dernier_mpich/2018.03.12.05h39m54s_c.txt" rel="noreferrer" target="_blank" moz-do-not-send="true">http://www.giref.ulaval.ca/~cm<wbr>pgiref/dernier_mpich/2018.03.<wbr>12.05h39m54s_c.txt</a>
                          <br>
                          <a href="http://www.giref.ulaval.ca/%7Ecmpgiref/dernier_mpich/2018.03.12.05h39m54s_m.txt" rel="noreferrer" target="_blank" moz-do-not-send="true">http://www.giref.ulaval.ca/~cm<wbr>pgiref/dernier_mpich/2018.03.<wbr>12.05h39m54s_m.txt</a>
                          <br>
                          <a href="http://www.giref.ulaval.ca/%7Ecmpgiref/dernier_mpich/2018.03.12.05h39m54s_mi.txt" rel="noreferrer" target="_blank" moz-do-not-send="true">http://www.giref.ulaval.ca/~cm<wbr>pgiref/dernier_mpich/2018.03.<wbr>12.05h39m54s_mi.txt</a>
                          <br>
                          <a href="http://www.giref.ulaval.ca/%7Ecmpgiref/dernier_mpich/2018.03.12.05h39m54s_openmpa_config.log" rel="noreferrer" target="_blank" moz-do-not-send="true">http://www.giref.ulaval.ca/~cm<wbr>pgiref/dernier_mpich/2018.03.<wbr>12.05h39m54s_openmpa_config.<wbr>log</a>
                          <br>
                          <a href="http://www.giref.ulaval.ca/%7Ecmpgiref/dernier_mpich/2018.03.12.05h39m54s_mpl_config.log" rel="noreferrer" target="_blank" moz-do-not-send="true">http://www.giref.ulaval.ca/~cm<wbr>pgiref/dernier_mpich/2018.03.<wbr>12.05h39m54s_mpl_config.log</a>
                          <br>
                          <a href="http://www.giref.ulaval.ca/%7Ecmpgiref/dernier_mpich/2018.03.12.05h39m54s_pm_hydra_config.log" rel="noreferrer" target="_blank" moz-do-not-send="true">http://www.giref.ulaval.ca/~cm<wbr>pgiref/dernier_mpich/2018.03.<wbr>12.05h39m54s_pm_hydra_config.<wbr>log</a>
                          <br>
                          <a href="http://www.giref.ulaval.ca/%7Ecmpgiref/dernier_mpich/2018.03.12.05h39m54s_pm_hydra_tools_topo_config.log" rel="noreferrer" target="_blank" moz-do-not-send="true">http://www.giref.ulaval.ca/~cm<wbr>pgiref/dernier_mpich/2018.03.<wbr>12.05h39m54s_pm_hydra_tools_<wbr>topo_config.log</a>
                          <br>
                          <a href="http://www.giref.ulaval.ca/%7Ecmpgiref/dernier_mpich/2018.03.12.05h39m54s_mpiexec_info.txt" rel="noreferrer" target="_blank" moz-do-not-send="true">http://www.giref.ulaval.ca/~cm<wbr>pgiref/dernier_mpich/2018.03.<wbr>12.05h39m54s_mpiexec_info.txt</a>
                          <br>
                          <br>
                          Petsc logs:<br>
                          <a href="http://www.giref.ulaval.ca/%7Ecmpgiref/dernier_mpich/2018.03.12.05h39m54s_configure.log" rel="noreferrer" target="_blank" moz-do-not-send="true">http://www.giref.ulaval.ca/~cm<wbr>pgiref/dernier_mpich/2018.03.<wbr>12.05h39m54s_configure.log</a>
                          <br>
                          <a href="http://www.giref.ulaval.ca/%7Ecmpgiref/dernier_mpich/2018.03.12.05h39m54s_make.log" rel="noreferrer" target="_blank" moz-do-not-send="true">http://www.giref.ulaval.ca/~cm<wbr>pgiref/dernier_mpich/2018.03.<wbr>12.05h39m54s_make.log</a>
                          <br>
                          <a href="http://www.giref.ulaval.ca/%7Ecmpgiref/dernier_mpich/2018.03.12.05h39m54s_default.log" rel="noreferrer" target="_blank" moz-do-not-send="true">http://www.giref.ulaval.ca/~cm<wbr>pgiref/dernier_mpich/2018.03.<wbr>12.05h39m54s_default.log</a>
                          <br>
                          <a href="http://www.giref.ulaval.ca/%7Ecmpgiref/dernier_mpich/2018.03.12.05h39m54s_RDict.log" rel="noreferrer" target="_blank" moz-do-not-send="true">http://www.giref.ulaval.ca/~cm<wbr>pgiref/dernier_mpich/2018.03.<wbr>12.05h39m54s_RDict.log</a>
                          <br>
                          <a href="http://www.giref.ulaval.ca/%7Ecmpgiref/dernier_mpich/2018.03.12.05h39m54s_CMakeLists.txt" rel="noreferrer" target="_blank" moz-do-not-send="true">http://www.giref.ulaval.ca/~cm<wbr>pgiref/dernier_mpich/2018.03.<wbr>12.05h39m54s_CMakeLists.txt</a>
                          <br>
                          <br>
                          <br>
                        </blockquote>
                        <br>
                      </blockquote>
                    </div>
                  </div>
                  ______________________________<wbr>_________________<br>
                  discuss mailing list     <a href="mailto:discuss@mpich.org" target="_blank" moz-do-not-send="true">discuss@mpich.org</a><br>
                  To manage subscription options or unsubscribe:<br>
                  <a href="https://lists.mpich.org/mailman/listinfo/discuss" rel="noreferrer" target="_blank" moz-do-not-send="true">https://lists.mpich.org/mailma<wbr>n/listinfo/discuss</a><br>
                </blockquote>
                <br>
              </blockquote>
            </div>
            <br>
            <br clear="all">
            <div><br>
            </div>
            -- <br>
            <div class="gmail_signature" data-smartmail="gmail_signature">
              <div dir="ltr">
                <div>
                  <div dir="ltr">
                    <div>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</div>
                    <div><br>
                    </div>
                    <div><a href="http://www.caam.rice.edu/%7Emk51/" target="_blank" moz-do-not-send="true">https://www.cse.buffalo.edu/~knepley/</a><br>
                    </div>
                  </div>
                </div>
              </div>
            </div>
          </div>
        </div>
      </blockquote>
      <br>
    </blockquote>
    <br>
  </body>
</html>