December 2014 - discuss - mailman3.cels.anl.gov

Re: [mpich-discuss] (no subject)
by Seo, Sangmin 04 Dec '14

04 Dec '14

Hi Chafik, You can find examples using MPI_GATHERV if you google it. Here are some I found: http://mpi.deino.net/mpi_functions/MPI_Gatherv.html http://www.mpi-forum.org/docs/mpi-1.1/mpi-11-html/node70.html Please use this mailing list to ask questions related to MPICH. For general questions, I would suggest to use other developer websites, e.g., http://stackoverflow.com/. Best, Sangmin On Dec 4, 2014, at 10:27 AM, Chafik sanaa <san.chafik(a)gmail.com<mailto:[email protected]>> wrote: hi I want a simple program with the function MPI_GATHERV _______________________________________________ discuss mailing list discuss(a)mpich.org<mailto:[email protected]> To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss _______________________________________________ discuss mailing list discuss(a)mpich.org To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

1 0

(no subject)
by Chafik sanaa 04 Dec '14

04 Dec '14

hi I want a simple program with the function MPI_GATHERV _______________________________________________ discuss mailing list discuss(a)mpich.org To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] Get faster calculation speed
by "Antonio J. Peña" 04 Dec '14

04 Dec '14

Hi Peter, MPICH is an implementation of a message-passing communications API. It can be used as a way of communicating your threads. This mailing list is intended to be a forum to discuss issues in the MPICH implementation. If you need further assistance on this topic, you may ask in a different forum, such as stackoverflow. Best, Antonio On 12/04/2014 07:14 AM, Peter Kroon wrote: > Hi, > > I have a process that consumes about 1% of the cpu and I've got 8 thread. > Is it possible to get a higher calculation speed with mpich by combing > the threads? > > Best, > Peter > > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss -- Antonio J. Peña Postdoctoral Appointee Mathematics and Computer Science Division Argonne National Laboratory 9700 South Cass Avenue, Bldg. 240, Of. 3148 Argonne, IL 60439-4847 apenya(a)mcs.anl.gov www.mcs.anl.gov/~apenya _______________________________________________ discuss mailing list discuss(a)mpich.org To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

1 0

Get faster calculation speed
by Peter Kroon 04 Dec '14

04 Dec '14

Hi, I have a process that consumes about 1% of the cpu and I've got 8 thread. Is it possible to get a higher calculation speed with mpich by combing the threads? Best, Peter _______________________________________________ discuss mailing list discuss(a)mpich.org To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] [mpich-ibm] Problem with MPICH3/OpenPA on IBM P755
by Victor Vysotskiy 04 Dec '14

04 Dec '14

Hi Rob, just in case if the MPICH team needs an account on P775 for testing/debugging purposes, we can easily arrange it. Just let us know and we will open access to you. We would love to have MPICH3 working on our "monster" machine. With best regards Victor. _______________________________________________ discuss mailing list discuss(a)mpich.org To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] MPICH Run Time error
by Balaji, Pavan 03 Dec '14

03 Dec '14

My guess is that mpi_hello is buggy. Can't really help without seeing a simple code that is failing. -- Pavan > On Dec 3, 2014, at 5:46 AM, vipan sharma <vipan.badyal(a)gmail.com> wrote: > > Dear Sir/Madam > Greetings of the day !! > I have installed the MPICH3 on RHEL5 with the steps give in the installation manual > inside the tar file of Mpich3 . > I am using a cluster of 3 systems with all the settings done as suggested by the manual , > and successfully able to run the example given inside the package i.e. Pi value calculation (cpi) > using the command > $mpiexec -n 6 -f machinefile ./cpi > > Now when I am running some other process or some other examples the following error > is getting generated about which I have no clues how to get rid of the error > > Here I am listing the programm and the command I am using and error generating > > NANO35 if my master node > > [user@NANO35 test]$ mpiexec -n 6 -f /home/user/machinefile ./mpi_hello > [mpiexec@NANO35] control_cb (pm/pmiserv/pmiserv_cb.c:200): assert (!closed) failed > [mpiexec@NANO35] HYDT_dmxu_poll_wait_for_event (tools/demux/demux_poll.c:76): callback returned error status > [mpiexec@NANO35] HYD_pmci_wait_for_completion (pm/pmiserv/pmiserv_pmci.c:198): error waiting for event > [mpiexec@NANO35] main (ui/mpich/mpiexec.c:344): process manager error waiting for completion > > Machinefile looks like > NANO35:2 > NANO36:2 > NANO37:2 > NANO38:2 > > regards > vipan Sharma > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss -- Pavan Balaji ✉️ http://www.mcs.anl.gov/~balaji _______________________________________________ discuss mailing list discuss(a)mpich.org To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] [mpich-ibm] Problem with MPICH3/OpenPA on IBM P755
by Rob Latham 03 Dec '14

03 Dec '14

On 12/03/2014 11:59 AM, Victor Vysotskiy wrote: > Hi Rob, > > thanks for the information and hint about GPFS provided. Just to be sure, do I understand correctly that MPICH3 is not officially supported on Power7/AIX? Is there any plan to support this configuration in the future? > We're working on it! In the mean time, if you find bugs, let us know. ==rob -- Rob Latham Mathematics and Computer Science Division Argonne National Lab, IL USA _______________________________________________ discuss mailing list discuss(a)mpich.org To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] [patch] fix bug in NEWMAD and MXM netmods
by Kuleshov Aleksey 03 Dec '14

03 Dec '14

_______________________________________________ discuss mailing list discuss(a)mpich.org To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] [mpich-ibm] Problem with MPICH3/OpenPA on IBM P755
by Victor Vysotskiy 03 Dec '14

03 Dec '14

Hi Rob, thanks for the information and hint about GPFS provided. Just to be sure, do I understand correctly that MPICH3 is not officially supported on Power7/AIX? Is there any plan to support this configuration in the future? With best regards, Victor. _______________________________________________ discuss mailing list discuss(a)mpich.org To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

1 0

MPICH 3.1.3 installs potentially broken libtool .la files
by Markus Geimer 03 Dec '14

03 Dec '14

Dear MPICH developers, I noticed that the installed '.la' files for MPICH's libraries include two linker path ('-L') specifications in the 'dependency_libs' line referring to the directory where MPICH was built: dependency_libs=' -L<builddir>/src/mpl -L<builddir>/src/openpa/src \ -lrt -lpthread' In the past, I ran into link failures with specifying non-existent directories when using the gold linker, though I'm currently not able to reproduce this with a minimal example. In any case, pointing to the build directory in an '.la' file generally seems like a bad idea. As a fix, I propose to only add the '-L' entries to 'external_ldflags' (Makefile.am, line 51) if external MPL and/or OpenPA libraries are used. Using the internal copies is already taken care of by adding the corresponding local '.la' files to 'pmpi_convenience_libs'. Thanks, Markus -- Dr. Markus Geimer Juelich Supercomputing Centre Institute for Advanced Simulation Forschungszentrum Juelich GmbH 52425 Juelich, Germany Phone: +49-2461-61-1773 Fax: +49-2461-61-6656 E-mail: m.geimer(a)fz-juelich.de WWW: http://www.fz-juelich.de/jsc/ _______________________________________________ discuss mailing list discuss(a)mpich.org To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

1 0