<html>
  <head>
    <meta content="text/html; charset=utf-8" http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    Hi,<br>
    <br>
    I tried valgrind in MPI but it aborts very early, with the error msg
    regarding PETSc initialize.<br>
    <br>
    I retry again, using a lower resolution.<br>
    <br>
    GAMG  works, but BoomerAMG and hypre doesn't. Increasing cpu too
    high (80) also cause it to hang. 60 works fine.<br>
    <br>
    My grid size is 98x169x169<br>
    <br>
    But when I increase the resolution, GAMG can't work again.<br>
    <br>
    I tried to increase the cpu no but it still doesn't work.<br>
    <br>
    Previously, using single z direction partition, it work using GAMG
    and hypre. So what could be the problem?<br>
    <pre class="moz-signature" cols="72">Thank you.

Yours sincerely,

TAY wee-beng</pre>
    <div class="moz-cite-prefix">On 25/12/2015 12:33 AM, Matthew Knepley
      wrote:<br>
    </div>
    <blockquote
cite="mid:CAMYG4GkqJQxb=RZRRM0Lv25KDsyGBTfV9cRE40Xit-HrfXP1DA@mail.gmail.com"
      type="cite">
      <div dir="ltr">It sounds like you have memory corruption in a
        different part of the code. Run in valgrind.
        <div><br>
        </div>
        <div>  Matt</div>
      </div>
      <div class="gmail_extra"><br>
        <div class="gmail_quote">On Thu, Dec 24, 2015 at 10:14 AM, TAY
          wee-beng <span dir="ltr"><<a moz-do-not-send="true"
              href="mailto:zonexo@gmail.com" target="_blank">zonexo@gmail.com</a>></span>
          wrote:<br>
          <blockquote class="gmail_quote" style="margin:0 0 0
            .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi,<br>
            <br>
            I have this strange error. I converted my CFD code from a z
            directon only partition to the yz direction partition. The
            code works fine but when I increase the cpu no, strange
            things happen when solving the Poisson eqn.<br>
            <br>
            I increase cpu no from 24 to 40.<br>
            <br>
            Sometimes it works, sometimes it doesn't. When it doesn't,
            it just hangs there with no output, or it gives the error
            below:<br>
            <br>
            Using MPI_Barrier during debug shows that it hangs at<br>
            <br>
            call KSPSolve(ksp,b_rhs,xx,ierr).<br>
            <br>
            I use hypre BoomerAMG and GAMG
            (-poisson_pc_gamg_agg_nsmooths 1 -poisson_pc_type gamg)<br>
            <br>
            <br>
            Why is this so random? Also how do I debug this type of
            problem.<br>
            <br>
            <br>
            [32]PETSC ERROR:
            ------------------------------------------------------------------------<br>
            [32]PETSC ERROR: Caught signal number 11 SEGV: Segmentation
            Violation, probably memory access out of range<br>
            [32]PETSC ERROR: Try option -start_in_debugger or
            -on_error_attach_debugger<br>
            [32]PETSC ERROR: or see <a moz-do-not-send="true"
              href="http://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind"
              rel="noreferrer" target="_blank">http://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind</a><br>
            [32]PETSC ERROR: or try <a moz-do-not-send="true"
              href="http://valgrind.org" rel="noreferrer"
              target="_blank">http://valgrind.org</a> on GNU/linux and
            Apple Mac OS X to find memory corruption errors<br>
            [32]PETSC ERROR: likely location of problem given in stack
            below<br>
            [32]PETSC ERROR: ---------------------  Stack Frames
            ------------------------------------<br>
            [32]PETSC ERROR: Note: The EXACT line numbers in the stack
            are not available,<br>
            [32]PETSC ERROR:       INSTEAD the line number of the start
            of the function<br>
            [32]PETSC ERROR:       is given.<br>
            [32]PETSC ERROR: [32] HYPRE_SetupXXX line 174
            /home/wtay/Codes/petsc-3.6.2/src/ksp/pc/impls/hypre/hypre.c<br>
            [32]PETSC ERROR: [32] PCSetUp_HYPRE line 122
            /home/wtay/Codes/petsc-3.6.2/src/ksp/pc/impls/hypre/hypre.c<br>
            [32]PETSC ERROR: [32] PCSetUp line 945
            /home/wtay/Codes/petsc-3.6.2/src/ksp/pc/interface/precon.c<br>
            [32]PETSC ERROR: [32] KSPSetUp line 247
            /home/wtay/Codes/petsc-3.6.2/src/ksp/ksp/interface/itfunc.c<br>
            [32]PETSC ERROR: [32] KSPSolve line 510
            /home/wtay/Codes/petsc-3.6.2/src/ksp/ksp/interface/itfunc.c<br>
            [32]PETSC ERROR: --------------------- Error Message
            --------------------------------------------------------------<br>
            [32]PETSC ERROR: Signal received<br>
            [32]PETSC ERROR: See <a moz-do-not-send="true"
              href="http://www.mcs.anl.gov/petsc/documentation/faq.html"
              rel="noreferrer" target="_blank">http://www.mcs.anl.gov/petsc/documentation/faq.html</a>
            for trouble shooting.<br>
            [32]PETSC ERROR: Petsc Release Version 3.6.2, Oct, 02, 2015<br>
            [32]PETSC ERROR: ./a.out on a petsc-3.6.2_shared_gnu_debug
            named n12-40 by wtay Thu Dec 24 17:01:51 2015<br>
            [32]PETSC ERROR: Configure options
            --with-mpi-dir=/opt/ud/openmpi-1.8.8/
            --download-fblaslapack=1 --with-debugging=1
            --download-hypre=1
            --prefix=/home/wtay/Lib/petsc-3.6.2_shared_gnu_debug
            --known-mpi-shared=1 --with-shared-libraries
            --with-fortran-interfaces=1<br>
            [32]PETSC ERROR: #1 User provided function() line 0 in 
            unknown file<br>
--------------------------------------------------------------------------<br>
            MPI_ABORT was invoked on rank 32 in communicator
            MPI_COMM_WORLD<br>
            with errorcode 59.<span class="HOEnZb"><font color="#888888"><br>
                <br>
                -- <br>
                Thank you.<br>
                <br>
                Yours sincerely,<br>
                <br>
                TAY wee-beng<br>
                <br>
              </font></span></blockquote>
        </div>
        <br>
        <br clear="all">
        <div><br>
        </div>
        -- <br>
        <div class="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>
          -- Norbert Wiener</div>
      </div>
    </blockquote>
    <br>
  </body>
</html>