November 2013 - discuss - mailman3.cels.anl.gov

Re: [mpich-discuss] problem with MPI_SCATTERV
by Jeff Hammond 28 Nov '13

28 Nov '13

That's still wrong. Please read the documentation on how this function must be used. Jeff Sent from my iPhone > On Nov 28, 2013, at 5:48 PM, Danyang Su <danyang.su(a)gmail.com> wrote: > > Hi Junchao, > > Thanks for pointing out this. It works after I changed the recvcnt value to the maximum number of elements in receive buffer (integer). > > Danyang > >> On 28/11/2013 3:13 PM, Junchao Zhang wrote: >> The second scounts in your code is suspicious. >> According to http://www.mcs.anl.gov/research/projects/mpi/www/www3/MPI_Scatterv.html, sendcnts and recvcnt are of different types. >> >> >> --Junchao Zhang >> >> >>> On Thu, Nov 28, 2013 at 2:29 PM, Danyang Su <danyang.su(a)gmail.com> wrote: >>> Hi All, >>> >>> I ran into a problem with MPI_SCATTERV. When the sendcounts (scounts in code) is the same for every process, it works fine, but if the sendcounts is different (e.g., 1, 2, 3, 4 for 4 processes, respectively), there will be error as follows: >>> >>> Fatal error in PMPI_Scatterv: Message truncated, error stack: >>> PMPI_Scatterv(376)................: MPI_Scatterv(sbuf=0000000000000000, scnts=0000000000E16CD0, displs=0000000000E16CA0, >>> MPI_INT, rbuf=0000000002BD0050, rcount=1, MPI_INT, root=0, MPI_COMM_WORLD) failed >>> MPIR_Scatterv_impl(187)...........: >>> MPIR_Scatterv(144)................: >>> MPIDI_CH3U_Receive_data_found(129): Message from rank 0 and tag 6 truncated; 16 bytes received but buffer size is 4 >>> >>> The code is simple as follows: >>> >>> call MPI_SCATTERV(ja_in, scounts, displs, MPI_INT, ja, scounts, MPI_INT, 0, comm , ierr) >>> >>> ja_in is allocated for process #0 and ja is allocated for every process. The size of ja_in and ja is big enough to hold large dataset. >>> >>> Thanks and regards, >>> >>> Danyang >>> >>> _______________________________________________ >>> discuss mailing list discuss(a)mpich.org >>> 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 > > _______________________________________________ > 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] problem with MPI_SCATTERV
by Danyang Su 28 Nov '13

28 Nov '13

Hi Junchao, Thanks for pointing out this. It works after I changed the recvcnt value to the maximum number of elements in receive buffer (integer). Danyang On 28/11/2013 3:13 PM, Junchao Zhang wrote: > The second scounts in your code is suspicious. > According to > http://www.mcs.anl.gov/research/projects/mpi/www/www3/MPI_Scatterv.html, > sendcnts and recvcnt are of different types. > > > --Junchao Zhang > > > On Thu, Nov 28, 2013 at 2:29 PM, Danyang Su <danyang.su(a)gmail.com > <mailto:[email protected]>> wrote: > > Hi All, > > I ran into a problem with MPI_SCATTERV. When the sendcounts > (scounts in code) is the same for every process, it works fine, > but if the sendcounts is different (e.g., 1, 2, 3, 4 for 4 > processes, respectively), there will be error as follows: > > Fatal error in PMPI_Scatterv: Message truncated, error stack: > PMPI_Scatterv(376)................: > MPI_Scatterv(sbuf=0000000000000000, scnts=0000000000E16CD0, > displs=0000000000E16CA0, > MPI_INT, rbuf=0000000002BD0050, rcount=1, MPI_INT, root=0, > MPI_COMM_WORLD) failed > MPIR_Scatterv_impl(187)...........: > MPIR_Scatterv(144)................: > MPIDI_CH3U_Receive_data_found(129): Message from rank 0 and tag 6 > truncated; 16 bytes received but buffer size is 4 > > The code is simple as follows: > > call MPI_SCATTERV(ja_in, *scounts*, displs, MPI_INT, ja, > *scounts*, MPI_INT, 0, comm , ierr) > > ja_in is allocated for process #0 and ja is allocated for every > process. The size of ja_in and ja is big enough to hold large dataset. > > Thanks and regards, > > Danyang > > _______________________________________________ > 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

Re: [mpich-discuss] problem with MPI_SCATTERV
by Junchao Zhang 28 Nov '13

28 Nov '13

The second scounts in your code is suspicious. According to http://www.mcs.anl.gov/research/projects/mpi/www/www3/MPI_Scatterv.html, sendcnts and recvcnt are of different types. --Junchao Zhang On Thu, Nov 28, 2013 at 2:29 PM, Danyang Su <danyang.su(a)gmail.com> wrote: > Hi All, > > I ran into a problem with MPI_SCATTERV. When the sendcounts (scounts in > code) is the same for every process, it works fine, but if the sendcounts > is different (e.g., 1, 2, 3, 4 for 4 processes, respectively), there will > be error as follows: > > Fatal error in PMPI_Scatterv: Message truncated, error stack: > PMPI_Scatterv(376)................: MPI_Scatterv(sbuf=0000000000000000, > scnts=0000000000E16CD0, displs=0000000000E16CA0, > MPI_INT, rbuf=0000000002BD0050, rcount=1, MPI_INT, root=0, > MPI_COMM_WORLD) failed > MPIR_Scatterv_impl(187)...........: > MPIR_Scatterv(144)................: > MPIDI_CH3U_Receive_data_found(129): Message from rank 0 and tag 6 > truncated; 16 bytes received but buffer size is 4 > > The code is simple as follows: > > call MPI_SCATTERV(ja_in, *scounts*, displs, MPI_INT, ja, *scounts*, > MPI_INT, 0, comm , ierr) > > ja_in is allocated for process #0 and ja is allocated for every process. > The size of ja_in and ja is big enough to hold large dataset. > > Thanks and regards, > > Danyang > > _______________________________________________ > 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] ubuntu mpich
by Wesley Bland 28 Nov '13

28 Nov '13

Or if you're in the wrong directory. > On Nov 28, 2013, at 7:05 AM, Nicolas Rosner <nrosner(a)gmail.com> wrote: > > Is ./examples.cpi a valid binary for your platform? (Try the `file' > command on it.) > > "No such file or directory" is, among other things, what you get when > attempting to execute non-executables. > > Cheers, > Nicolás > > > > > On Thu, Nov 28, 2013 at 8:45 AM, Soheil Hooshdaran > <shooshdaran577(a)gmail.com> wrote: >> Dear sir/madam >> Hello. >> I am running Ubuntu 12.0.4 LTS (my computer is dual boot with win XP SP3 and >> ubuntu). I can swear that I have run the example 'cpi' in the package >> 'mpich-3.0.4'. but now, see the result of executting >> >>> soheil@soheil-desktop:~/mpich-3.0.4$ mpiexec -n 4 ./examples.cpi >>> [proxy:0:0@soheil-desktop] HYDU_create_process >>> (./utils/launch/launch.c:75): execvp error on file ./examples.cpi (No such >>> file or directory) >>> [proxy:0:0@soheil-desktop] HYDU_create_process >>> (./utils/launch/launch.c:75): execvp error on file ./examples.cpi (No such >>> file or directory) >>> [proxy:0:0@soheil-desktop] HYDU_create_process >>> (./utils/launch/launch.c:75): execvp error on file ./examples.cpi (No such >>> file or directory) >>> [proxy:0:0@soheil-desktop] HYDU_create_process >>> (./utils/launch/launch.c:75): execvp error on file ./examples.cpi (No such >>> file or directory) >>> >>> =================================================================================== >>> = BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES >>> = EXIT CODE: 255 >>> = CLEANING UP REMAINING PROCESSES >>> = YOU CAN IGNORE THE BELOW CLEANUP MESSAGES >>> >>> =================================================================================== >>> soheil@soheil-desktop:~/mpich-3.0.4$ >> >> >> Also, 'whiching' mpicc and mpiexec both return /user/local/bin/mpicc and >> /user/local/bin/mpiexec, respectively. >> >> I mean I have installed mpich before and sat the PATH variable. >> >> I need mpich on my ubuntu. Please help. >> >> >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> 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

problem with MPI_SCATTERV
by Danyang Su 28 Nov '13

28 Nov '13

Hi All, I ran into a problem with MPI_SCATTERV. When the sendcounts (scounts in code) is the same for every process, it works fine, but if the sendcounts is different (e.g., 1, 2, 3, 4 for 4 processes, respectively), there will be error as follows: Fatal error in PMPI_Scatterv: Message truncated, error stack: PMPI_Scatterv(376)................: MPI_Scatterv(sbuf=0000000000000000, scnts=0000000000E16CD0, displs=0000000000E16CA0, MPI_INT, rbuf=0000000002BD0050, rcount=1, MPI_INT, root=0, MPI_COMM_WORLD) failed MPIR_Scatterv_impl(187)...........: MPIR_Scatterv(144)................: MPIDI_CH3U_Receive_data_found(129): Message from rank 0 and tag 6 truncated; 16 bytes received but buffer size is 4 The code is simple as follows: call MPI_SCATTERV(ja_in, *scounts*, displs, MPI_INT, ja, *scounts*, MPI_INT, 0, comm , ierr) ja_in is allocated for process #0 and ja is allocated for every process. The size of ja_in and ja is big enough to hold large dataset. Thanks and regards, Danyang

1 0

Re: [mpich-discuss] Blocking MPI recieve
by Huiwei Lu 28 Nov '13

28 Nov '13

See this following tutorial for more details: https://computing.llnl.gov/tutorials/mpi/ Huiwei Lu http://www.mcs.anl.gov/~huiweilu/ On Nov 28, 2013, at 12:34 PM, Soheil Hooshdaran <shooshdaran577(a)gmail.com> wrote: > Hello. > In > > if (rank == 0) > - > - > MPI_Send(x); > - > - > - > else > - > - > - > MPI_Recv(y) > - > - > - > > Does the code that comes after MPI_Recv() start as soon as MPI_Recv() finishes its execution? > _______________________________________________ > 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] apt-get install mpich
by Nicolas Rosner 28 Nov '13

28 Nov '13

If you needed sudo before nano, you also need sudo before apt-get: $ sudo apt-get install mpich On Thu, Nov 28, 2013 at 3:28 PM, Soheil Hooshdaran <shooshdaran577(a)gmail.com> wrote: > apt-get install mpich

1 0

Blocking MPI recieve
by Soheil Hooshdaran 28 Nov '13

28 Nov '13

Hello. In if (rank == 0) - - MPI_Send(x); - - - else - - - MPI_Recv(y) - - - Does the code that comes after MPI_Recv() start as soon as MPI_Recv() finishes its execution?

1 0

apt-get install mpich
by Soheil Hooshdaran 28 Nov '13

28 Nov '13

Hi. As you told me I tyoed: soheil@soheil-desktop:/media/TOSHIBA$ sudo nano /etc/apt/sources.list [sudo] password for soheil: soheil@soheil-desktop:/media/TOSHIBA$ apt-get install mpich But got: E: Could not open lock file /var/lib/dpkg/lock - open (13: Permission denied) E: Unable to lock the administration directory (/var/lib/dpkg/), are you root? soheil@soheil-desktop:/media/TOSHIBA$ sudo apt-get install mpich Reading package lists... Done Building dependency tree Reading state information... Done Package mpich is not available, but is referred to by another package. This may mean that the package is missing, has been obsoleted, or is only available from another source E: Package 'mpich' has no installation candidate soheil@soheil-desktop:/media/TOSHIBA$ What do I do now?

1 0

Re: [mpich-discuss] discuss Digest, Vol 13, Issue 30
by Marcelo Soares Souza 28 Nov '13

28 Nov '13

> > On Ubuntu/Debian, the package is still called “mpich2”. In my repository i call mpich, and was updated to 3.0.4 (last stable) >> soheil@soheil-desktop:/media/TOSHIBA$ apt-get install mpich For all command always use root, after editing sources.list * sudo apt-get update * sudo apt-get install mpich

1 0