<html><head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
  </head>
  <body>
    <p>Hi all.</p>
    <p>I have asked Thibault (author or this report on HDF5 <a href="https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.mcs.anl.gov%2Fpipermail%2Fpetsc-users%2F2021-July%2F044045.html&data=04%7C01%7Cmatteo.semplice%40uninsubria.it%7C0443f8463c604f1122fb08d9485ac463%7C9252ed8bdffc401c86ca6237da9991fa%7C0%7C0%7C637620376946153189%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=w%2BAY%2FEZ7pPUintp77eEN1vGJVB3R0VX1Pzp9U0xV4E4%3D&reserved=0" rel="noreferrer" target="_blank">https://lists.mcs.anl.gov/pipermail/petsc-users/2021-July/044045.html</a>
      some days before mine) to run my MWE and it does not work for him
      either.</p>
    <p>Further, I have tried on another machine of mine with
      --download-hdf5 --download-mpich and still it is not working.</p>
    <p>A detailed report follows at the end of this message.</p>
    <p>I am wondering if something is wrong/incompatible with the HDF5
      version of VecView, at least when the Vec is associated with a
      DMDA. Of course it might just be that I didn't manage to write a
      correct xdmf, but I can't spot the mistake...<br>
    </p>
    <p>I am of course available to run tests in order to find/fix this
      problem.</p>
    <p>Best</p>
    <p>    Matteo<br>
    </p>
    <div class="moz-cite-prefix">On 16/07/21 12:27, Matteo Semplice
      wrote:<br>
    </div>
    <blockquote type="cite" cite="mid:6c443aac-e9c6-0704-beb5-05afa8c38798@uninsubria.it">
      <br>
      Il 15/07/21 17:44, Matteo Semplice ha scritto:
      <br>
      <blockquote type="cite">Hi.
        <br>
        <br>
        When I write (HDF5 viewer) a vector associated to a DMDA with 1
        dof, the output is independent of the number of cpus used.
        <br>
        <br>
        However, for a DMDA with dof=2, the output seems to be correct
        when I run on 1 or 2 cpus, but is scrambled when I run with 4
        cpus. Judging from the ranges of the data, each field gets
        written to the correct part, and its the data witin the field
        that is scrambled. Here's my MWE:
        <br>
        <br>
        #include <petscversion.h>
        <br>
        #include <petscdmda.h>
        <br>
        #include <petscviewer.h>
        <br>
        #include <petscsys.h>
        <br>
        #include <petscviewerhdf5.h>
        <br>
        <br>
        int main(int argc, char **argv) {
        <br>
        <br>
          PetscErrorCode ierr;
        <br>
          ierr = PetscInitialize(&argc,&argv,(char*)0,help);
        CHKERRQ(ierr);
        <br>
          PetscInt Nx=11;
        <br>
          PetscInt Ny=11;
        <br>
          PetscScalar dx = 1.0 / (Nx-1);
        <br>
          PetscScalar dy = 1.0 / (Ny-1);
        <br>
          DM dmda;
        <br>
          ierr = DMDACreate2d(PETSC_COMM_WORLD,
        <br>
                              DM_BOUNDARY_NONE,DM_BOUNDARY_NONE,
        <br>
                              DMDA_STENCIL_STAR,
        <br>
                              Nx,Ny, //global dim
        <br>
                              PETSC_DECIDE,PETSC_DECIDE, //n proc on
        each dim
        <br>
                              2,1, //dof, stencil width
        <br>
                              NULL, NULL, //n nodes per direction on
        each cpu
        <br>
                              &dmda);      CHKERRQ(ierr);
        <br>
          ierr = DMSetFromOptions(dmda); CHKERRQ(ierr);
        <br>
          ierr = DMSetUp(dmda); CHKERRQ(ierr); CHKERRQ(ierr);
        <br>
          ierr = DMDASetUniformCoordinates(dmda, 0.0, 1.0, 0.0, 1.0,
        0.0, 1.0); CHKERRQ(ierr);
        <br>
          ierr = DMDASetFieldName(dmda,0,"s"); CHKERRQ(ierr);
        <br>
          ierr = DMDASetFieldName(dmda,1,"c"); CHKERRQ(ierr);
        <br>
          DMDALocalInfo daInfo;
        <br>
          ierr = DMDAGetLocalInfo(dmda,&daInfo); CHKERRQ(ierr);
        <br>
          IS *is;
        <br>
          DM *daField;
        <br>
          ierr = DMCreateFieldDecomposition(dmda,NULL, NULL, &is,
        &daField); CHKERRQ(ierr);
        <br>
          Vec U0;
        <br>
          ierr = DMCreateGlobalVector(dmda,&U0); CHKERRQ(ierr);
        <br>
        <br>
          //Initial data
        <br>
          typedef struct{ PetscScalar s,c;} data_type;
        <br>
          data_type **u;
        <br>
          ierr = DMDAVecGetArray(dmda,U0,&u); CHKERRQ(ierr);
        <br>
          for (PetscInt j=daInfo.ys; j<daInfo.ys+daInfo.ym; j++){
        <br>
            PetscScalar y = j*dy;
        <br>
            for (PetscInt i=daInfo.xs; i<daInfo.xs+daInfo.xm; i++){
        <br>
              PetscScalar x = i*dx;
        <br>
              u[j][i].s = x+2.*y;
        <br>
              u[j][i].c = 10. + 2.*x*x+y*y;
        <br>
            }
        <br>
          }
        <br>
          ierr = DMDAVecRestoreArray(dmda,U0,&u); CHKERRQ(ierr);
        <br>
        <br>
          PetscViewer viewer;
        <br>
          ierr =
PetscViewerHDF5Open(PETSC_COMM_WORLD,"solutionSC.hdf5",FILE_MODE_WRITE,&viewer);
        CHKERRQ(ierr);
        <br>
          Vec uField;
        <br>
          ierr = VecGetSubVector(U0,is[0],&uField); CHKERRQ(ierr);
        <br>
          PetscObjectSetName((PetscObject) uField, "S");
        <br>
          ierr = VecView(uField,viewer); CHKERRQ(ierr);
        <br>
          ierr = VecRestoreSubVector(U0,is[0],&uField);
        CHKERRQ(ierr);
        <br>
          ierr = VecGetSubVector(U0,is[1],&uField); CHKERRQ(ierr);
        <br>
          PetscObjectSetName((PetscObject) uField, "C");
        <br>
          ierr = VecView(uField,viewer); CHKERRQ(ierr);
        <br>
          ierr = VecRestoreSubVector(U0,is[1],&uField);
        CHKERRQ(ierr);
        <br>
          ierr = PetscViewerDestroy(&viewer); CHKERRQ(ierr);
        <br>
        <br>
          ierr = PetscFinalize();
        <br>
          return ierr;
        <br>
        }
        <br>
        <br>
        and my xdmf file
        <br>
        <br>
        <?xml version="1.0" ?>
        <br>
        <Xdmf
xmlns:xi=<a class="moz-txt-link-rfc2396E" href="https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.w3.org%2F2001%2FXInclude&amp;data=04%7C01%7Cmatteo.semplice%40uninsubria.it%7C7c270ed0c49c4f8d950708d948444e1c%7C9252ed8bdffc401c86ca6237da9991fa%7C0%7C0%7C637620280470927505%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=q45En4ULjQX6H%2F1ZgzUxgKmDk7Y7jK0K2IuWDHpr4HM%3D&amp;reserved=0">"https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.w3.org%2F2001%2FXInclude&amp;data=04%7C01%7Cmatteo.semplice%40uninsubria.it%7C7c270ed0c49c4f8d950708d948444e1c%7C9252ed8bdffc401c86ca6237da9991fa%7C0%7C0%7C637620280470927505%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=q45En4ULjQX6H%2F1ZgzUxgKmDk7Y7jK0K2IuWDHpr4HM%3D&amp;reserved=0"</a>
        Version="2.0">
        <br>
          <Domain>
        <br>
            <Grid GridType="Collection" CollectionType="Temporal">
        <br>
              <Time TimeType="List">
        <br>
                <DataItem Dimensions="1">1.0</DataItem>
        <br>
              </Time>
        <br>
              <Grid GridType="Uniform" Name="domain">
        <br>
                <Topology TopologyType="2DCoRectMesh" Dimensions="11
        11"/>
        <br>
                <Geometry GeometryType="ORIGIN_DXDY">
        <br>
                  <DataItem Format="XML" NumberType="Float"
        Dimensions="2">0.0 0.0</DataItem>
        <br>
                  <DataItem Format="XML" NumberType="Float"
        Dimensions="2">0.1 0.1</DataItem>
        <br>
                </Geometry>
        <br>
                <Attribute Name="S" Center="Node"
        AttributeType="Scalar">
        <br>
                  <DataItem Format="HDF" Precision="8" Dimensions="11
        11">solutionSC.hdf5:/S</DataItem>
        <br>
                </Attribute>
        <br>
                <Attribute Name="C" Center="Node"
        AttributeType="Scalar">
        <br>
                  <DataItem Format="HDF" Precision="8" Dimensions="11
        11">solutionSC.hdf5:/C</DataItem>
        <br>
                </Attribute>
        <br>
              </Grid>
        <br>
            </Grid>
        <br>
          </Domain>
        <br>
        </Xdmf>
        <br>
        <br>
        Steps to reprduce: run code and open the xdmf with paraview. If
        the code was run with 1,2 or 3 cpus, the data are correct
        (except the plane xy has become the plane yz), but with 4 cpus
        the data are scrambled.
        <br>
        <br>
        Does anyone have any insight?
        <br>
        <br>
        (I am using Petsc Release Version 3.14.2, but I can compile a
        newer one if you think it's important.)
        <br>
      </blockquote>
      <br>
      Hi,
      <br>
      <br>
          I have a small update on this issue.
      <br>
      <br>
      First, it is still here with version 3.15.2.
      <br>
      <br>
      Secondly, I have run the code under valgrind and
      <br>
      <br>
      - for 1 or 2 processes, I get no errors
      <br>
      <br>
      - for 4 processes, 3 out of 4, trigger the following
      <br>
      <br>
      ==25921== Conditional jump or move depends on uninitialised
      value(s)
      <br>
      ==25921==    at 0xB3D6259: ??? (in
      /usr/lib/x86_64-linux-gnu/openmpi/lib/openmpi3/mca_fcoll_two_phase.so)
      <br>
      ==25921==    by 0xB3D85C8: mca_fcoll_two_phase_file_write_all (in
/usr/lib/x86_64-linux-gnu/openmpi/lib/openmpi3/mca_fcoll_two_phase.so)
      <br>
      ==25921==    by 0xAAEB29B: mca_common_ompio_file_write_at_all (in
/usr/lib/x86_64-linux-gnu/openmpi/lib/libmca_common_ompio.so.41.9.0)
      <br>
      ==25921==    by 0xB316605: mca_io_ompio_file_write_at_all (in
      /usr/lib/x86_64-linux-gnu/openmpi/lib/openmpi3/mca_io_ompio.so)
      <br>
      ==25921==    by 0x73C7FE7: PMPI_File_write_at_all (in
      /usr/lib/x86_64-linux-gnu/openmpi/lib/libmpi.so.40.10.3)
      <br>
      ==25921==    by 0x69E8700: H5FD__mpio_write (H5FDmpio.c:1466)
      <br>
      ==25921==    by 0x670D6EB: H5FD_write (H5FDint.c:248)
      <br>
      ==25921==    by 0x66DA0D3: H5F__accum_write (H5Faccum.c:826)
      <br>
      ==25921==    by 0x684F091: H5PB_write (H5PB.c:1031)
      <br>
      ==25921==    by 0x66E8055: H5F_shared_block_write (H5Fio.c:205)
      <br>
      ==25921==    by 0x6674538: H5D__chunk_collective_fill
      (H5Dchunk.c:5064)
      <br>
      ==25921==    by 0x6674538: H5D__chunk_allocate (H5Dchunk.c:4736)
      <br>
      ==25921==    by 0x668C839: H5D__init_storage (H5Dint.c:2473)
      <br>
      ==25921==  Uninitialised value was created by a heap allocation
      <br>
      ==25921==    at 0x483577F: malloc (vg_replace_malloc.c:299)
      <br>
      ==25921==    by 0xB3D6155: ??? (in
      /usr/lib/x86_64-linux-gnu/openmpi/lib/openmpi3/mca_fcoll_two_phase.so)
      <br>
      ==25921==    by 0xB3D85C8: mca_fcoll_two_phase_file_write_all (in
/usr/lib/x86_64-linux-gnu/openmpi/lib/openmpi3/mca_fcoll_two_phase.so)
      <br>
      ==25921==    by 0xAAEB29B: mca_common_ompio_file_write_at_all (in
/usr/lib/x86_64-linux-gnu/openmpi/lib/libmca_common_ompio.so.41.9.0)
      <br>
      ==25921==    by 0xB316605: mca_io_ompio_file_write_at_all (in
      /usr/lib/x86_64-linux-gnu/openmpi/lib/openmpi3/mca_io_ompio.so)
      <br>
      ==25921==    by 0x73C7FE7: PMPI_File_write_at_all (in
      /usr/lib/x86_64-linux-gnu/openmpi/lib/libmpi.so.40.10.3)
      <br>
      ==25921==    by 0x69E8700: H5FD__mpio_write (H5FDmpio.c:1466)
      <br>
      ==25921==    by 0x670D6EB: H5FD_write (H5FDint.c:248)
      <br>
      ==25921==    by 0x66DA0D3: H5F__accum_write (H5Faccum.c:826)
      <br>
      ==25921==    by 0x684F091: H5PB_write (H5PB.c:1031)
      <br>
      ==25921==    by 0x66E8055: H5F_shared_block_write (H5Fio.c:205)
      <br>
      ==25921==    by 0x6674538: H5D__chunk_collective_fill
      (H5Dchunk.c:5064)
      <br>
      ==25921==    by 0x6674538: H5D__chunk_allocate (H5Dchunk.c:4736)
      <br>
      <br>
      Does anyone have any hint on what might be causing this?
      <br>
      <br>
      Is this the "buggy MPI-IO" that Matt was mentioning in
<a class="moz-txt-link-freetext" href="https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.mcs.anl.gov%2Fpipermail%2Fpetsc-users%2F2021-July%2F044138.html&amp;data=04%7C01%7Cmatteo.semplice%40uninsubria.it%7C7c270ed0c49c4f8d950708d948444e1c%7C9252ed8bdffc401c86ca6237da9991fa%7C0%7C0%7C637620280470927505%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=gPAbClDgJ1toQxzVVnoRCrgPBNR2tjw%2BGfdrxv%2FwVmY%3D&amp;reserved=0">https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.mcs.anl.gov%2Fpipermail%2Fpetsc-users%2F2021-July%2F044138.html&amp;data=04%7C01%7Cmatteo.semplice%40uninsubria.it%7C7c270ed0c49c4f8d950708d948444e1c%7C9252ed8bdffc401c86ca6237da9991fa%7C0%7C0%7C637620280470927505%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=gPAbClDgJ1toQxzVVnoRCrgPBNR2tjw%2BGfdrxv%2FwVmY%3D&amp;reserved=0</a>?<br>
      <br>
      I am using the release branch (commit c548142fde) and I have
      configured with --download-hdf5; configure finds the installed
      openmpi 3.1.3 from Debian buster. The relevant lines from
      configure.log are
      <br>
      <br>
      MPI:
      <br>
        Version:  3
      <br>
        Mpiexec: mpiexec --oversubscribe
      <br>
        OMPI_VERSION: 3.1.3
      <br>
      hdf5:
      <br>
        Version:  1.12.0
      <br>
        Includes: -I/home/matteo/software/petsc/opt/include
      <br>
        Library:  -Wl,-rpath,/home/matteo/software/petsc/opt/lib
      -L/home/matteo/software/petsc/opt/lib -lhdf5hl_fortran
      -lhdf5_fortran -lhdf5_hl -lhdf5
      <br>
    </blockquote>
    <p>Update 1: on a different machine, I have compiled petsc (release
      branch) with --download-hdf5 and --download-mpich and I have tried
      3d HDF5 output at the end of my simulation. All's fine for 1 or 2
      CPUs, but the output is funny for more CPUs. The smooth solution
      gives rise to an output that renders like little bricks, as if the
      data were written doing the 3 nested loops in the wrong order.</p>
    <p>Update 2: Thibault was kind enough to compile and run my MWE on
      his setup and he gets a crash related to the VecView with the HDF5
      viewer. Here's the report that he sent me.<br>
    </p>
    <div>On 21/07/21 10:59, Thibault Bridel-Bertomeu wrote:<br>
    </div>
    Hi Matteo, 
    <div><br>
    </div>
    <div>I ran your test, and actually it does not give me garbage for a
      number of processes greater than 1, it straight-up crashes ...</div>
    <div>Here is the error log for 2 processes :</div>
    <div><br>
    </div>
    <div>
      <p><span>Compiled with Petsc Development GIT revision:
          v3.14.4-671-g707297fd510<span>  </span>GIT Date: 2021-02-24
          22:50:05 +0000</span></p>
      <p><span>[1]PETSC ERROR:
          ------------------------------------------------------------------------</span></p>
      <p><span>[1]PETSC ERROR: Caught signal number 11 SEGV:
          Segmentation Violation, probably memory access out of range</span></p>
      <p><span>[1]PETSC ERROR: Try option -start_in_debugger or
          -on_error_attach_debugger</span></p>
      <p><span>[1]PETSC ERROR: or see <a href="https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.mcs.anl.gov%2Fpetsc%2Fdocumentation%2Ffaq.html%23valgrind&data=04%7C01%7Cmatteo.semplice%40uninsubria.it%7C4046642d31454b13ad8708d94c299c2c%7C9252ed8bdffc401c86ca6237da9991fa%7C0%7C0%7C637624563866796549%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=ZFBJWmDy6QUnBGYC2mHWBz%2F%2Bx%2BYan2Pw4ub%2FhR8PvK4%3D&reserved=0" target="_blank">https://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind</a></span></p>
      <p><span>[1]PETSC ERROR: or try <a href="https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fvalgrind.org%2F&data=04%7C01%7Cmatteo.semplice%40uninsubria.it%7C4046642d31454b13ad8708d94c299c2c%7C9252ed8bdffc401c86ca6237da9991fa%7C0%7C0%7C637624563866806506%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=5aITh8yt1ihtXQzuhBzG5vYy8ZeQhDq%2Bhj3aLMGjEvc%3D&reserved=0" target="_blank">http://valgrind.org</a> on GNU/linux and
          Apple Mac OS X to find memory corruption errors</span></p>
      <p><span>[1]PETSC ERROR: likely location of problem given in stack
          below</span></p>
      <p><span>[1]PETSC ERROR: ---------------------<span>  </span>Stack
          Frames ------------------------------------</span></p>
      <p><span>[1]PETSC ERROR: Note: The EXACT line numbers in the stack
          are not available,</span></p>
      <p><span>[1]PETSC ERROR: <span>      </span>INSTEAD the line
          number of the start of the function</span></p>
      <p><span>[1]PETSC ERROR: <span>      </span>is given.</span></p>
      <p><span>[1]PETSC ERROR: [1] H5Dcreate2 line 716
          /ccc/work/cont001/ocre/bridelbert/04-PETSC/src/vec/vec/impls/mpi/pdvec.c</span></p>
      <p><span>[1]PETSC ERROR: [1] VecView_MPI_HDF5 line 622
          /ccc/work/cont001/ocre/bridelbert/04-PETSC/src/vec/vec/impls/mpi/pdvec.c</span></p>
      <p><span>[1]PETSC ERROR: [1] VecView_MPI line 815
          /ccc/work/cont001/ocre/bridelbert/04-PETSC/src/vec/vec/impls/mpi/pdvec.c</span></p>
      <p><span>[1]PETSC ERROR: [1] VecView line 580
/ccc/work/cont001/ocre/bridelbert/04-PETSC/src/vec/vec/interface/vector.c</span></p>
      <p><span>[1]PETSC ERROR: --------------------- Error Message
          --------------------------------------------------------------</span></p>
      <p><span>[1]PETSC ERROR: Signal received</span></p>
      <p><span>[1]PETSC ERROR: See <a href="https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.mcs.anl.gov%2Fpetsc%2Fdocumentation%2Ffaq.html&data=04%7C01%7Cmatteo.semplice%40uninsubria.it%7C4046642d31454b13ad8708d94c299c2c%7C9252ed8bdffc401c86ca6237da9991fa%7C0%7C0%7C637624563866816465%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=Obo%2BDTzUQK%2BAzcBn7rmfrZwiFIhxv%2B9SAF5BQPVLCAg%3D&reserved=0" target="_blank">https://www.mcs.anl.gov/petsc/documentation/faq.html</a>
          for trouble shooting.</span></p>
      <p><span>[1]PETSC ERROR: Petsc Development GIT revision:
          v3.14.4-671-g707297fd510<span>  </span>GIT Date: 2021-02-24
          22:50:05 +0000</span></p>
      <p><span>[1]PETSC ERROR:
          /ccc/work/cont001/ocre/bridelbert/MWE_HDF5_Output/testHDF5 on
          a<span>  </span>named r1login by bridelbert Wed Jul 21
          10:57:11 2021</span></p>
      <p><span>[1]PETSC ERROR: Configure options --with-clean=1
          --prefix=/ccc/work/cont001/ocre/bridelbert/04-PETSC/build_uns3D_inti
          --with-make-np=8 --with-windows-graphics=0 --with-debugging=1
          --download-mpich-shared=0 --with-x=0 --with-pthread=0
          --with-valgrind=0 --PETSC_ARCH=INTI_UNS3D
--with-fc=/ccc/products/openmpi-2.0.4/gcc--8.3.0/default/bin/mpifort
          --with-cc=/ccc/products/openmpi-2.0.4/gcc--8.3.0/default/bin/mpicc
--with-cxx=/ccc/products/openmpi-2.0.4/gcc--8.3.0/default/bin/mpicxx
          --with-openmp=0
          --download-sowing=/ccc/work/cont001/ocre/bridelbert/v1.1.26-p1.tar.gz
--download-metis=/ccc/work/cont001/ocre/bridelbert/git.metis.tar.gz
--download-parmetis=/ccc/work/cont001/ocre/bridelbert/git.parmetis.tar.gz
--download-fblaslapack=/ccc/work/cont001/ocre/bridelbert/git.fblaslapack.tar.gz
          --with-cmake-dir=/ccc/products/cmake-3.13.3/system/default
--download-hdf5=/ccc/work/cont001/ocre/bridelbert/hdf5-1.12.0.tar.bz2
          --download-zlib=/ccc/work/cont001/ocre/bridelbert/zlib-1.2.11.tar.gz</span></p>
      <p><span>[1]PETSC ERROR: #1 User provided function() line 0 in<span> 
          </span>unknown file</span></p>
      <p><span>[1]PETSC ERROR: Checking the memory for corruption.</span></p>
      <p><span>--------------------------------------------------------------------------</span></p>
      <p><span>MPI_ABORT was invoked on rank 1 in communicator
          MPI_COMM_WORLD</span></p>
      <p><span>with errorcode 50176059.</span></p>
      <p><span></span><br>
      </p>
      <p><span>NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI
          processes.</span></p>
      <p><span>You may or may not see output from other processes,
          depending on</span></p>
      <p><span>exactly when Open MPI kills them.</span></p>
      <p><span>--------------------------------------------------------------------------</span></p>
      <p><span>[0]PETSC ERROR:
          ------------------------------------------------------------------------</span></p>
      <p><span>[0]PETSC ERROR: Caught signal number 15 Terminate: Some
          process (or the batch system) has told this process to end</span></p>
      <p><span>[0]PETSC ERROR: Try option -start_in_debugger or
          -on_error_attach_debugger</span></p>
      <p><span>[0]PETSC ERROR: or see <a href="https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.mcs.anl.gov%2Fpetsc%2Fdocumentation%2Ffaq.html%23valgrind&data=04%7C01%7Cmatteo.semplice%40uninsubria.it%7C4046642d31454b13ad8708d94c299c2c%7C9252ed8bdffc401c86ca6237da9991fa%7C0%7C0%7C637624563866816465%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=EzrdlWi7y%2FWU7LsRuqGd%2BAp%2B17bdX8bOxKbszA15k5E%3D&reserved=0" target="_blank">https://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind</a></span></p>
      <p><span>[0]PETSC ERROR: or try <a href="https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fvalgrind.org%2F&data=04%7C01%7Cmatteo.semplice%40uninsubria.it%7C4046642d31454b13ad8708d94c299c2c%7C9252ed8bdffc401c86ca6237da9991fa%7C0%7C0%7C637624563866826422%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=o%2BFJ7bq2dp5bokp%2BKQj1efXC7Dqfbl0yBjHi2xobvug%3D&reserved=0" target="_blank">http://valgrind.org</a> on GNU/linux and
          Apple Mac OS X to find memory corruption errors</span></p>
      <p><span>[0]PETSC ERROR: likely location of problem given in stack
          below</span></p>
      <p><span>[0]PETSC ERROR: ---------------------<span>  </span>Stack
          Frames ------------------------------------</span></p>
      <p><span>[0]PETSC ERROR: Note: The EXACT line numbers in the stack
          are not available,</span></p>
      <p><span>[0]PETSC ERROR: <span>      </span>INSTEAD the line
          number of the start of the function</span></p>
      <p><span>[0]PETSC ERROR: <span>      </span>is given.</span></p>
      <p><span>[0]PETSC ERROR: [0] H5Dcreate2 line 716
          /ccc/work/cont001/ocre/bridelbert/04-PETSC/src/vec/vec/impls/mpi/pdvec.c</span></p>
      <p><span>[0]PETSC ERROR: [0] VecView_MPI_HDF5 line 622
          /ccc/work/cont001/ocre/bridelbert/04-PETSC/src/vec/vec/impls/mpi/pdvec.c</span></p>
      <p><span>[0]PETSC ERROR: [0] VecView_MPI line 815
          /ccc/work/cont001/ocre/bridelbert/04-PETSC/src/vec/vec/impls/mpi/pdvec.c</span></p>
      <p><span>[0]PETSC ERROR: [0] VecView line 580
/ccc/work/cont001/ocre/bridelbert/04-PETSC/src/vec/vec/interface/vector.c</span></p>
      <p><span>[0]PETSC ERROR: --------------------- Error Message
          --------------------------------------------------------------</span></p>
      <p><span>[0]PETSC ERROR: Signal received</span></p>
      <p><span>[0]PETSC ERROR: See <a href="https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.mcs.anl.gov%2Fpetsc%2Fdocumentation%2Ffaq.html&data=04%7C01%7Cmatteo.semplice%40uninsubria.it%7C4046642d31454b13ad8708d94c299c2c%7C9252ed8bdffc401c86ca6237da9991fa%7C0%7C0%7C637624563866826422%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=TZ3qXboAnlDtWc6%2Bh5V175dPVzqCUhTTyFdCtm3J7lM%3D&reserved=0" target="_blank">https://www.mcs.anl.gov/petsc/documentation/faq.html</a>
          for trouble shooting.</span></p>
      <p><span>[0]PETSC ERROR: Petsc Development GIT revision:
          v3.14.4-671-g707297fd510<span>  </span>GIT Date: 2021-02-24
          22:50:05 +0000</span></p>
      <p><span>[0]PETSC ERROR:
          /ccc/work/cont001/ocre/bridelbert/MWE_HDF5_Output/testHDF5 on
          a<span>  </span>named r1login by bridelbert Wed Jul 21
          10:57:11 2021</span></p>
      <p><span>[0]PETSC ERROR: Configure options --with-clean=1
          --prefix=/ccc/work/cont001/ocre/bridelbert/04-PETSC/build_uns3D_inti
          --with-make-np=8 --with-windows-graphics=0 --with-debugging=1
          --download-mpich-shared=0 --with-x=0 --with-pthread=0
          --with-valgrind=0 --PETSC_ARCH=INTI_UNS3D
--with-fc=/ccc/products/openmpi-2.0.4/gcc--8.3.0/default/bin/mpifort
          --with-cc=/ccc/products/openmpi-2.0.4/gcc--8.3.0/default/bin/mpicc
--with-cxx=/ccc/products/openmpi-2.0.4/gcc--8.3.0/default/bin/mpicxx
          --with-openmp=0
          --download-sowing=/ccc/work/cont001/ocre/bridelbert/v1.1.26-p1.tar.gz
--download-metis=/ccc/work/cont001/ocre/bridelbert/git.metis.tar.gz
--download-parmetis=/ccc/work/cont001/ocre/bridelbert/git.parmetis.tar.gz
--download-fblaslapack=/ccc/work/cont001/ocre/bridelbert/git.fblaslapack.tar.gz
          --with-cmake-dir=/ccc/products/cmake-3.13.3/system/default
--download-hdf5=/ccc/work/cont001/ocre/bridelbert/hdf5-1.12.0.tar.bz2
          --download-zlib=/ccc/work/cont001/ocre/bridelbert/zlib-1.2.11.tar.gz</span></p>
      <p><span>[0]PETSC ERROR: #1 User provided function() line 0 in<span> 
          </span>unknown file</span></p>
      <p><span>[r1login:24498] 1 more process has sent help message
          help-mpi-api.txt / mpi-abort</span></p>
      <p><span>[r1login:24498] Set MCA parameter
          "orte_base_help_aggregate" to 0 to see all help / error
          messages</span></p>
    </div>
    <br>
    <div>I am starting to wonder if the PETSc configure script installs
      HDF5 with MPI correctly at all ...</div>
    <div><br>
    </div>
    <div>
      <div>Here is my conf :</div>
      <div><br>
      </div>
      <div>
        <p class="gmail-p1"><span class="gmail-s1">Compilers:</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>C Compiler:
            <span class="gmail-Apple-converted-space">        </span>/ccc/products/openmpi-2.0.4/gcc--8.3.0/default/bin/mpicc<span class="gmail-Apple-converted-space">  </span>-fPIC -Wall
            -Wwrite-strings -Wno-strict-aliasing -Wno-unknown-pragmas
            -fstack-protector -fvisibility=hidden -g3</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">    </span>Version:
            gcc (GCC) 8.3.0</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>C++
            Compiler: <span class="gmail-Apple-converted-space">       
            </span>/ccc/products/openmpi-2.0.4/gcc--8.3.0/default/bin/mpicxx<span class="gmail-Apple-converted-space">  </span>-Wall
            -Wwrite-strings -Wno-strict-aliasing -Wno-unknown-pragmas
            -fstack-protector -fvisibility=hidden -g<span class="gmail-Apple-converted-space">  </span>-fPIC</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">    </span>Version:
            g++ (GCC) 8.3.0</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Fortran
            Compiler: <span class="gmail-Apple-converted-space">       
            </span>/ccc/products/openmpi-2.0.4/gcc--8.3.0/default/bin/mpifort<span class="gmail-Apple-converted-space">  </span>-fPIC -Wall
            -ffree-line-length-0 -Wno-unused-dummy-argument -g</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">    </span>Version:
            GNU Fortran (GCC) 8.3.0</span></p>
        <p class="gmail-p1"><span class="gmail-s1">Linkers:</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Shared
            linker: <span class="gmail-Apple-converted-space">  </span>/ccc/products/openmpi-2.0.4/gcc--8.3.0/default/bin/mpicc<span class="gmail-Apple-converted-space">  </span>-shared<span class="gmail-Apple-converted-space">  </span>-fPIC -Wall
            -Wwrite-strings -Wno-strict-aliasing -Wno-unknown-pragmas
            -fstack-protector -fvisibility=hidden -g3</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Dynamic
            linker: <span class="gmail-Apple-converted-space">  </span>/ccc/products/openmpi-2.0.4/gcc--8.3.0/default/bin/mpicc<span class="gmail-Apple-converted-space">  </span>-shared<span class="gmail-Apple-converted-space">  </span>-fPIC -Wall
            -Wwrite-strings -Wno-strict-aliasing -Wno-unknown-pragmas
            -fstack-protector -fvisibility=hidden -g3</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Libraries
            linked against: <span class="gmail-Apple-converted-space"> 
            </span>-lquadmath -lstdc++ -ldl</span></p>
        <p class="gmail-p1"><span class="gmail-s1">BlasLapack:</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Library:<span class="gmail-Apple-converted-space">  </span>-Wl,-rpath,/ccc/work/cont001/ocre/bridelbert/04-PETSC/build_uns3D_inti/lib
-L/ccc/work/cont001/ocre/bridelbert/04-PETSC/build_uns3D_inti/lib
            -lflapack -lfblas</span></p>
      </div>
      <div>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>uses 4 byte
            integers</span></p>
        <p class="gmail-p1"><span class="gmail-s1">MPI:</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Version:<span class="gmail-Apple-converted-space">  </span>3</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Mpiexec:
            /ccc/products/openmpi-2.0.4/gcc--8.3.0/default/bin/mpiexec</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>OMPI_VERSION:
            2.0.4</span></p>
        <p class="gmail-p1"><span class="gmail-s1">fblaslapack:</span></p>
        <p class="gmail-p1"><span class="gmail-s1">zlib:</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Version:<span class="gmail-Apple-converted-space">  </span>1.2.11</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Includes:
            -I/ccc/work/cont001/ocre/bridelbert/04-PETSC/build_uns3D_inti/include</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Library:<span class="gmail-Apple-converted-space">  </span>-Wl,-rpath,/ccc/work/cont001/ocre/bridelbert/04-PETSC/build_uns3D_inti/lib
-L/ccc/work/cont001/ocre/bridelbert/04-PETSC/build_uns3D_inti/lib -lz</span></p>
        <p class="gmail-p1"><span class="gmail-s1">hdf5:</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Version:<span class="gmail-Apple-converted-space">  </span>1.12.0</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Includes:
            -I/ccc/work/cont001/ocre/bridelbert/04-PETSC/build_uns3D_inti/include</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Library:<span class="gmail-Apple-converted-space">  </span>-Wl,-rpath,/ccc/work/cont001/ocre/bridelbert/04-PETSC/build_uns3D_inti/lib
-L/ccc/work/cont001/ocre/bridelbert/04-PETSC/build_uns3D_inti/lib
            -lhdf5hl_fortran -lhdf5_fortran -lhdf5_hl -lhdf5</span></p>
        <p class="gmail-p1"><span class="gmail-s1">cmake:</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Version:<span class="gmail-Apple-converted-space">  </span>3.13.3</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>/ccc/products/cmake-3.13.3/system/default/bin/cmake</span></p>
        <p class="gmail-p1"><span class="gmail-s1">metis:</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Version:<span class="gmail-Apple-converted-space">  </span>5.1.0</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Includes:
            -I/ccc/work/cont001/ocre/bridelbert/04-PETSC/build_uns3D_inti/include</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Library:<span class="gmail-Apple-converted-space">  </span>-Wl,-rpath,/ccc/work/cont001/ocre/bridelbert/04-PETSC/build_uns3D_inti/lib
-L/ccc/work/cont001/ocre/bridelbert/04-PETSC/build_uns3D_inti/lib
            -lmetis</span></p>
        <p class="gmail-p1"><span class="gmail-s1">parmetis:</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Version:<span class="gmail-Apple-converted-space">  </span>4.0.3</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Includes:
            -I/ccc/work/cont001/ocre/bridelbert/04-PETSC/build_uns3D_inti/include</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Library:<span class="gmail-Apple-converted-space">  </span>-Wl,-rpath,/ccc/work/cont001/ocre/bridelbert/04-PETSC/build_uns3D_inti/lib
-L/ccc/work/cont001/ocre/bridelbert/04-PETSC/build_uns3D_inti/lib
            -lparmetis</span></p>
        <p class="gmail-p1"><span class="gmail-s1">regex:</span></p>
        <p class="gmail-p1"><span class="gmail-s1">sowing:</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Version:<span class="gmail-Apple-converted-space">  </span>1.1.26</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>/ccc/work/cont001/ocre/bridelbert/04-PETSC/INTI_UNS3D/bin/bfort</span></p>
        <p class="gmail-p1"><span class="gmail-s1"><span class="gmail-Apple-converted-space">  </span>Language
            used to compile PETSc: C</span></p>
      </div>
      <br>
      <div>Please don't hesitate to ask if you need something else from
        me !</div>
      <div><br>
      </div>
      <div>Cheers, </div>
      <div>Thibault</div>
    </div>
  </body>
</html>