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

Re: [mpich-discuss] Error "hydra requires VA_ARGS support", RHEL 5.10, icc
by Balaji, Pavan 28 Feb '14

28 Feb '14

That’s surprising, given that it’s incorrect. :-) I have a corrected patch pending. — Pavan On Feb 28, 2014, at 1:19 AM, Бакиров, Медер <bakirov(a)almetico.com> wrote: > I successfully compiled and installed latest nightly release without any errors. > > Thanks! > > With the best regards, > Meder Bakirov > > С уважением, > Бакиров Медер > > > On Thu, Feb 27, 2014 at 9:23 PM, Lu, Huiwei <huiweilu(a)mcs.anl.gov> wrote: >> Done reviewing. >> Tested with CentOS 6.5, icc version 13.1.3. >> $ export CFLAGS='-O3 -static-intel' >> $ export CXXFLAGS='-O3 -static-intel' >> $ export FFLAGS='-O3 -static-intel' >> $ ./configure && make >> >> — >> Huiwei >> >> On Feb 27, 2014, at 7:36 AM, Balaji, Pavan <balaji(a)anl.gov> wrote: >> >>> >>> I’ve pushed a review branch in for Huiwei. Once he signs off on it, I’ll commit it in an create a “nightly" tarball. I’ll send out an email once that’s done. It should be in today. >>> >>> — Pavan >>> >>> On Feb 27, 2014, at 4:36 AM, Бакиров, Медер <bakirov(a)almetico.com> wrote: >>> >>>> On Thu, Feb 27, 2014 at 12:59 PM, Balaji, Pavan <balaji(a)anl.gov> wrote: >>>>> Alternatively, you can wait till this is reviewed internally and committed to mpich/master. Then it'll show up in the nightly "release-like" tarballs. >>>> >>>> >>>> Thank you very much! I think, I will wait till this is reviewed and >>>> committed. Where can I follow a progress on this? >>>> >>>> With the best regards, >>>> Meder >>>> _______________________________________________ >>>> 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 > _______________________________________________ > 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] MPID_nem_tcp_init(377)... gethostbyname failed, mac (errno 1)
by Luis Cebamanos 28 Feb '14

28 Feb '14

Hi, I have also tried the following, with similar result: $ mpiexec -n 1 -host localhost ./hello [mpiexec@mac] HYDU_sock_is_local (utils/sock/sock.c:532): gethostbyname error on mac: Unknown host [mpiexec@mac] main (ui/mpich/mpiexec.c:291): unable to check if localhost is local thought it could add some more info. Cheers, Luis On 28 Feb 2014, at 12:25, Luis Cebamanos <l.cebamanos(a)epcc.ed.ac.uk> wrote: > Hi all, > > I have installed MPICH2 in a Mac 10.9.2 using homebrew. The installation process went OK, no problems were reported. Then I created a small hello world example just to test the MPI: > > #include <stdio.h> > #include "mpi.h" > > int main (argc, argv) > int argc; > char *argv[]; > { > int rank, size; > > MPI_Init (&argc, &argv);/* starts MPI */ > MPI_Comm_rank (MPI_COMM_WORLD, &rank);/* get current process id */ > MPI_Comm_size (MPI_COMM_WORLD, &size);/* get number of processes */ > > printf( "Hello world from process %d of %d\n", rank, size ); > > MPI_Finalize(); > return 0; > } > > compile it ($mpicc -o hello hello.c), all went fine too; and finally run it ($mpiexec -n 2 ./hello). However, that is the problem reported by mpiexec: > > $ mpiexec -n 2 ./hello > Fatal error in MPI_Init: Other MPI error, error stack: > MPIR_Init_thread(467)..............: > MPID_Init(177).....................: channel initialization failed > MPIDI_CH3_Init(70).................: > MPID_nem_init(319).................: > MPID_nem_tcp_init(171).............: > MPID_nem_tcp_get_business_card(418): > MPID_nem_tcp_init(377).............: gethostbyname failed, mac (errno 1) > Fatal error in MPI_Init: Other MPI error, error stack: > MPIR_Init_thread(467)..............: > MPID_Init(177).....................: channel initialization failed > MPIDI_CH3_Init(70).................: > MPID_nem_init(319).................: > MPID_nem_tcp_init(171).............: > MPID_nem_tcp_get_business_card(418): > MPID_nem_tcp_init(377).............: gethostbyname failed, mac (errno 1) > > =================================================================================== > = BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES > = PID 653 RUNNING AT mac > = EXIT CODE: 1 > = CLEANING UP REMAINING PROCESSES > = YOU CAN IGNORE THE BELOW CLEANUP MESSAGES > =================================================================================== > > The version installed: > $ mpichversion > MPICH Version: 3.1 > MPICH Release date: Thu Feb 20 11:41:13 CST 2014 > MPICH Device: ch3:nemesis > MPICH configure: --disable-dependency-tracking --disable-silent-rules --prefix=/usr/local/Cellar/mpich2/3.1 --mandir=/usr/local/Cellar/mpich2/3.1/share/man > MPICH CC: clang -O2 > MPICH CXX: clang++ > MPICH F77: /usr/local/bin/gfortran -O2 > MPICH FC: /usr/local/bin/gfortran > > Does anyone have any idea of why I am getting this error? The hostname command is also working fine, so I do NOT understand this problem. Furthermore, the idea is to run MPICH2 on my mac, not in a cluster so it should be straightforward to pick the hostname. My /etc/hosts shows the following: > $ cat /etc/hosts > ## > # Host Database > # > # localhost is used to configure the loopback interface > # when the system is booting. Do not change this entry. > ## > 127.0.0.1 localhost > 255.255.255.255 broadcasthost > ::1 localhost > fe80::1%lo0 localhost > > Any ideas? Any help, suggestions or workarounds would be much appreciated. > > Best regards, > LuisThe University of Edinburgh is a charitable body, registered in > Scotland, with registration number SC005336. > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.

1 0

MPID_nem_tcp_init(377)... gethostbyname failed, mac (errno 1)
by Luis Cebamanos 28 Feb '14

28 Feb '14

Hi all, I have installed MPICH2 in a Mac 10.9.2 using homebrew. The installation process went OK, no problems were reported. Then I created a small hello world example just to test the MPI: #include <stdio.h> #include "mpi.h" int main (argc, argv) int argc; char *argv[]; { int rank, size; MPI_Init (&argc, &argv);/* starts MPI */ MPI_Comm_rank (MPI_COMM_WORLD, &rank);/* get current process id */ MPI_Comm_size (MPI_COMM_WORLD, &size);/* get number of processes */ printf( "Hello world from process %d of %d\n", rank, size ); MPI_Finalize(); return 0; } compile it ($mpicc -o hello hello.c), all went fine too; and finally run it ($mpiexec -n 2 ./hello). However, that is the problem reported by mpiexec: $ mpiexec -n 2 ./hello Fatal error in MPI_Init: Other MPI error, error stack: MPIR_Init_thread(467)..............: MPID_Init(177).....................: channel initialization failed MPIDI_CH3_Init(70).................: MPID_nem_init(319).................: MPID_nem_tcp_init(171).............: MPID_nem_tcp_get_business_card(418): MPID_nem_tcp_init(377).............: gethostbyname failed, mac (errno 1) Fatal error in MPI_Init: Other MPI error, error stack: MPIR_Init_thread(467)..............: MPID_Init(177).....................: channel initialization failed MPIDI_CH3_Init(70).................: MPID_nem_init(319).................: MPID_nem_tcp_init(171).............: MPID_nem_tcp_get_business_card(418): MPID_nem_tcp_init(377).............: gethostbyname failed, mac (errno 1) =================================================================================== = BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES = PID 653 RUNNING AT mac = EXIT CODE: 1 = CLEANING UP REMAINING PROCESSES = YOU CAN IGNORE THE BELOW CLEANUP MESSAGES =================================================================================== The version installed: $ mpichversion MPICH Version: 3.1 MPICH Release date: Thu Feb 20 11:41:13 CST 2014 MPICH Device: ch3:nemesis MPICH configure: --disable-dependency-tracking --disable-silent-rules --prefix=/usr/local/Cellar/mpich2/3.1 --mandir=/usr/local/Cellar/mpich2/3.1/share/man MPICH CC: clang -O2 MPICH CXX: clang++ MPICH F77: /usr/local/bin/gfortran -O2 MPICH FC: /usr/local/bin/gfortran Does anyone have any idea of why I am getting this error? The hostname command is also working fine, so I do NOT understand this problem. Furthermore, the idea is to run MPICH2 on my mac, not in a cluster so it should be straightforward to pick the hostname. My /etc/hosts shows the following: $ cat /etc/hosts ## # Host Database # # localhost is used to configure the loopback interface # when the system is booting. Do not change this entry. ## 127.0.0.1 localhost 255.255.255.255 broadcasthost ::1 localhost fe80::1%lo0 localhost Any ideas? Any help, suggestions or workarounds would be much appreciated. Best regards, Luis The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.

1 0

Re: [mpich-discuss] Problems with running make
by Reuti 28 Feb '14

28 Feb '14

Hi, Am 28.02.2014 um 09:22 schrieb Ron Palmer: > I am struggling with running make on MPICH, V3.1. I believe that ./configure went well but I do not understand what caused the errors in make. I thought it may be associated with the fortran compile but I have F77 installed (though somewhat confused whether fortran 90 is also installed). The README file suggested that I send an email to this list with nominated four text files attached. You can try to skip f90: $ ./configure --disable-fc ... -- Reuti > The end of the m.txt file is included here for easy access, full details are attached... > > CC src/mpi/topo/nhb_alltoallw.lo > CC src/binding/f90/create_f90_int.lo > CC src/binding/f90/create_f90_real.lo > src/binding/f90/create_f90_real.c: In function 'PMPI_Type_create_f90_real': > src/binding/f90/create_f90_real.c:73: error: expected expression before ',' token > src/binding/f90/create_f90_real.c:74: error: expected expression before ',' token > make[2]: *** [src/binding/f90/create_f90_real.lo] Error 1 > make[2]: Leaving directory `/inv/mpich-3.1' > make[1]: *** [all-recursive] Error 1 > make[1]: Leaving directory `/inv/mpich-3.1' > make: *** [all] Error 2 > > I look forward to any suggestions any of you may have. > Regards, Ron > > > <mpi_logs.tar.gz>_______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0

Problems with running make
by Ron Palmer 28 Feb '14

28 Feb '14

I am struggling with running make on MPICH, V3.1. I believe that ./configure went well but I do not understand what caused the errors in make. I thought it may be associated with the fortran compile but I have F77 installed (though somewhat confused whether fortran 90 is also installed). The README file suggested that I send an email to this list with nominated four text files attached. The end of the m.txt file is included here for easy access, full details are attached... CC src/mpi/topo/nhb_alltoallw.lo CC src/binding/f90/create_f90_int.lo CC src/binding/f90/create_f90_real.lo src/binding/f90/create_f90_real.c: In function 'PMPI_Type_create_f90_real': src/binding/f90/create_f90_real.c:73: error: expected expression before ',' token src/binding/f90/create_f90_real.c:74: error: expected expression before ',' token make[2]: *** [src/binding/f90/create_f90_real.lo] Error 1 make[2]: Leaving directory `/inv/mpich-3.1' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/inv/mpich-3.1' make: *** [all] Error 2 I look forward to any suggestions any of you may have. Regards, Ron

1 0

Re: [mpich-discuss] Error "hydra requires VA_ARGS support", RHEL 5.10, icc
by Бакиров, Медер 28 Feb '14

28 Feb '14

I successfully compiled and installed latest nightly release without any errors. Thanks! With the best regards, Meder Bakirov С уважением, Бакиров Медер On Thu, Feb 27, 2014 at 9:23 PM, Lu, Huiwei <huiweilu(a)mcs.anl.gov> wrote: > Done reviewing. > Tested with CentOS 6.5, icc version 13.1.3. > $ export CFLAGS='-O3 -static-intel' > $ export CXXFLAGS='-O3 -static-intel' > $ export FFLAGS='-O3 -static-intel' > $ ./configure && make > > — > Huiwei > > On Feb 27, 2014, at 7:36 AM, Balaji, Pavan <balaji(a)anl.gov> wrote: > >> >> I’ve pushed a review branch in for Huiwei. Once he signs off on it, I’ll commit it in an create a “nightly" tarball. I’ll send out an email once that’s done. It should be in today. >> >> — Pavan >> >> On Feb 27, 2014, at 4:36 AM, Бакиров, Медер <bakirov(a)almetico.com> wrote: >> >>> On Thu, Feb 27, 2014 at 12:59 PM, Balaji, Pavan <balaji(a)anl.gov> wrote: >>>> Alternatively, you can wait till this is reviewed internally and committed to mpich/master. Then it'll show up in the nightly "release-like" tarballs. >>> >>> >>> Thank you very much! I think, I will wait till this is reviewed and >>> committed. Where can I follow a progress on this? >>> >>> With the best regards, >>> Meder >>> _______________________________________________ >>> 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

Fwd: The bug in handling 2D arrays
by Lu, Huiwei 27 Feb '14

27 Feb '14

Forward to the mailing list. — Huiwei Begin forwarded message: From: Soheil Hooshdaran <shooshdaran577(a)gmail.com<mailto:[email protected]>> Subject: Re: The bug in handling 2D arrays Date: February 27, 2014 at 2:12:21 PM CST To: "Lu, Huiwei" <huiweilu(a)mcs.anl.gov<mailto:[email protected]>> Oh, ha!ha!. sorry.Attached On Thu, Feb 27, 2014 at 10:18 PM, Lu, Huiwei <huiweilu(a)mcs.anl.gov<mailto:[email protected]>> wrote: Hi, Soheil, you may have forgot to attach the source file. — Huiwei Lu On Feb 27, 2014, at 12:31 PM, Soheil Hooshdaran <shooshdaran577(a)gmail.com<mailto:[email protected]>> wrote: > > Hello. > Please note the instruction at line 105 (the MPI_Alltoall call) and the comment on the line 108. > You only need to run the program once. > The program takes 3 arguments from the command line: > 1-How many numbers does each processor produce locally > 2-The range bounds of numbers to produces randomly(like , say, 2 to 20, entered as 2 20) > > so to run, e.g, type mpirun -n 4 ./a.out 4 0 15 > > Note that the number of processors should be a power of 2 > > I personally think that maybe the reason for this bug is that the index and the size of each dimension are ignored and that only the base address and the displacement is assumed. > > I read in Intel developer's manual that you need 4 parameters to access an element of a multi-dimensional array. I wonder if you can apply this. -- قال المصطفی علیه السلام: ‌«إرحموا ثلاثا عزیزاً، قوم الذل و غنی قوم افتقر و عالماً یلعب به الجُهّال» --------- به سما و به بروجش به هبوط و به عروجش نفسی همتک بادم نفسی من هلپندم نفسی آتش سوزان نفسی سیل گریزان ز چه اصلم ز چه فصلم به چه بازار خرندم نفسی فوق طباقم نفسی شام و عراقم نفسی غرق فراقم نفسی راز تو رندم نفسی همره ماهم نفسی مست الهم نفسی یوسف چاهم نفسی جمله گزندم نفسی رهزن و غولم نفسی تند و ملولم نفسی زین دو برونم که بر آن بام بلندم ------------------------------------------------------------------------------------------------ Being nice to someone you hate doesn't mean you're fake. It means you're mature enough to tolerate your hate towards them.

1 0

The bug in handling 2D arrays
by Soheil Hooshdaran 27 Feb '14

27 Feb '14

Hello. Please note the instruction at line 105 (the *MPI_Alltoall* call) and the comment on the line 108. You only need to run the program once. The program takes 3 arguments from the command line: 1-How many numbers does each processor produce locally 2-The range bounds of numbers to produces randomly(like , say, 2 to 20, entered as 2 20) so to run, e.g, type mpirun -n 4 ./a.out 4 0 15 Note that the number of processors should be a power of 2 I personally think that maybe the reason for this bug is that the *index*and the *size* of each dimension are ignored and that only the *base address* and the *displacement* is assumed. I read in Intel developer's manual that you need 4 parameters to access an element of a multi-dimensional array. I wonder if you can apply this.

1 0

Re: [mpich-discuss] Question
by "Antonio J. Peña" 27 Feb '14

27 Feb '14

In a homogeneous execution all MPI processes should have the same environment variables. If that's not your case, maybe you could consider just reading it by means of a call to getenv and using MPI sends/receives/collectives/one-sided. You can afterwards setenv it if that's what you want. Antonio On 02/27/2014 11:23 AM, Ramesh Vinayagam wrote: > Hi, > > How do I send enviroment variables to a spawned process spawned using > MPI_Comm_spawn. I tried to use MPI_Info, but it doesn't seem to work. > Does anyone have a example that I could look at. > > > Please let me know, > > Thanks > Ramesh > > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss -- Antonio J. Peï¿½a 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

1 0

Question
by Ramesh Vinayagam 27 Feb '14

27 Feb '14

Hi, How do I send enviroment variables to a spawned process spawned using MPI_Comm_spawn. I tried to use MPI_Info, but it doesn't seem to work. Does anyone have a example that I could look at. Please let me know, Thanks Ramesh

1 0