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

Re: [mpich-discuss] Very long configuration time
by Yida Wang 21 May '14

21 May '14

Thanks for your reply, Antonio! Yes, I was using NFS before, am trying on local disk now. BTW, what will a normal time be? On Wed, May 21, 2014 at 2:20 PM, "Antonio J. Peña" <apenya(a)mcs.anl.gov>wrote: > > Hi YW, > > Although 8h looks like a very long time to me, it's likely due to a > network filesystem. If that's the case, configuring MPICH in a local > filesystem (e.g. /temp, /tmp, /scratch) should go back to normal times. > > Antonio > > > > On 05/21/2014 01:17 PM, Yida Wang wrote: > > Hi Guys, > > I would like to try out the feature of talking to Intel-MIC in MPICH2. > Therefore I downloaded mpich-3.1 and configured it as > >> "./configure --host=x86_64-k1om-linux CC=icc CXX=icpc F77=ifort >> FC=ifort CFLAGS=-mmic CXXFLAGS=-mmic FFLAGS=-mmic FCFLAGS=-mmic >> LDFLAGS=-mmic" > > > However, it took a very long time (i.e. 8 hours) for me to finish the > configuration. Is it normal? Notice that my system doesn't have Infiniband > but 10GigE connection. However, I don't think this would affect the > configuration time. > > Thanks, > YW > > > _______________________________________________ > 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 60439-4847apenya(a)mcs.anl.govwww.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

Re: [mpich-discuss] Very long configuration time
by "Antonio J. Peña" 21 May '14

21 May '14

Hi YW, Although 8h looks like a very long time to me, it's likely due to a network filesystem. If that's the case, configuring MPICH in a local filesystem (e.g. /temp, /tmp, /scratch) should go back to normal times. Antonio On 05/21/2014 01:17 PM, Yida Wang wrote: > Hi Guys, > > I would like to try out the feature of talking to Intel-MIC in MPICH2. > Therefore I downloaded mpich-3.1 and configured it as > > "./configure --host=x86_64-k1om-linux CC=icc CXX=icpc F77=ifort > FC=ifort CFLAGS=-mmic CXXFLAGS=-mmic FFLAGS=-mmic FCFLAGS=-mmic > LDFLAGS=-mmic" > > > However, it took a very long time (i.e. 8 hours) for me to finish the > configuration. Is it normal? Notice that my system doesn't have > Infiniband but 10GigE connection. However, I don't think this would > affect the configuration time. > > Thanks, > YW > > > _______________________________________________ > 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

Very long configuration time
by Yida Wang 21 May '14

21 May '14

Hi Guys, I would like to try out the feature of talking to Intel-MIC in MPICH2. Therefore I downloaded mpich-3.1 and configured it as > "./configure --host=x86_64-k1om-linux CC=icc CXX=icpc F77=ifort > FC=ifort CFLAGS=-mmic CXXFLAGS=-mmic FFLAGS=-mmic FCFLAGS=-mmic > LDFLAGS=-mmic" However, it took a very long time (i.e. 8 hours) for me to finish the configuration. Is it normal? Notice that my system doesn't have Infiniband but 10GigE connection. However, I don't think this would affect the configuration time. Thanks, YW

1 0

Re: [mpich-discuss] bug in MPI_File_write_all?
by CANELA-XANDRI Oriol 21 May '14

21 May '14

Ok, many thanks for your help. Oriol -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. -----Original Message----- From: discuss-bounces(a)mpich.org [mailto:[email protected]] On Behalf Of Rob Latham Sent: 21 May 2014 14:06 To: discuss(a)mpich.org Subject: Re: [mpich-discuss] bug in MPI_File_write_all? On 05/21/2014 05:50 AM, CANELA-XANDRI Oriol wrote: > Ok, so it seems a problem with parallel access to this file system. Other network file systems like NFS, for instance, should not be a problem, right? > > Do you know if there is some method just for checking that this kind of problems will not arise in a particular system (with an 'unknown' file system)? > you can use NFS if you must but NFS is also not a true parallel file system. One could simply write to a local file system for a start. The PVFS parallel file system is another option. It's free and easy to deploy. Many sites have Lustre, though it is a bit more complicated. ==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] Fault tolerance MPICH2 hydra vs MPICH3.0.4 hydra
by Wesley Bland 21 May '14

21 May '14

Yes, this is expected. Fault tolerance is an experimental feature and as such is not implemented in all devices. It is currently only compatible with the TCP device. Thanks, Wesley > On May 21, 2014, at 8:40 AM, Anatoly G <anatolyrishon(a)gmail.com> wrote: > > > Thank you, Pavan. > I tried MPICH 3.1, it works good with Fault tolerance. > One more question: > If I execute my simulation (previous mail) with MPICH 3.1 compiled with --with-device=ch3:sock, > I see not stable fault tolerance. Sometimes whole system crashes, but sometimes not. > If I use default configuration flags (without --with-device=ch3:sock) whole system is stable. > > > Is this expected behavior? > > > Regards, > Anatoly. > > > > On Mon, May 19, 2014 at 5:03 PM, Balaji, Pavan<balaji(a)anl.gov >>wrote: >> >> Hydra should be compatible between different versions of mpich (or mpich derivatives). However, there’s always a possibility that there was a bug in mpich-3.0.4’s hydra that was fixed in mpich-3.1. So we recommend using the latest version. >> >> — Pavan >> >> On May 19, 2014, at 1:15 AM, Anatoly G <anatolyrishon(a)gmail.com >>> wrote: >> >>> Hi Wesley. >>> Thank you very much for quick response. >>> I executed your's code. Master can't finish it's execution. It stalled on MPI_Wait on iteration 7. >>> >>> But if I use MPICH2 hydra, Master process will finish executing by reporting number of times on failure of slave process. >>> >>> Can you please advice if it's safe to make hybrid system of build with MPICH3.0.4, but using MPICH2 hydra? >>> Or may be any other solution. >>> Does MPICH 3.0.4 include all MPICH2 hydra functionality? >>> May be my configuration of MPICH 3.0.4 is wrong? >>> >>> Regards, >>> Anatoly. >>> >>> >>> On Sun, May 18, 2014 at 8:40 PM, Wesley Bland <wbland(a)anl.gov >>> wrote: >>> Hi Anatoly, >>> >>> I think the problem may be the way that you're aborting. MPICH catches the system abort call and kills the entire application when it's called. Instead, I suggest using MPI_Abort(MPI_COMM_WORLD, 1); That's what I use in my tests and it works fine. It also seemed to work for your code when I tried. I'll attach my modified version of your code. I switched it to C since I happened to have C++ support disabled on my local install, but that shouldn't change anything. >>> >>> Thanks, >>> Wesley >>> >>> >>> On Sun, May 18, 2014 at 5:18 AM, Anatoly G <anatolyrishon(a)gmail.com >>> wrote: >>> Dear MPICH2, >>> Can you please help me with understanding Fault Tolerance in MPICH3.0.4 >>> I have a simple MPI program: >>> Master calls MPI_Irecv + MPI_Wait in loop. >>> Single slave: calls MPI_Send x 5 times, then calls abort. >>> >>> When I execute program with MPICH2 hydra I get multiple times Master process prints about fail in slave. In MPICH3 hydra I get a single message about fail of slave and then Master process enters to endless wait for next Irecv completion. >>> In both cases I compiled program with MPICH3.0.4 >>> >>> In other words, with MPICH2 hydra each Irecv completes (even if slave died before execution of Irecv) but in MPICH3 hydra not. Causes MPI_Irecv endless wait. >>> >>> If I compile same program with MPICH2 and use MPICH2 hydra, I get the same result as compiling with MPICH3.0.4 and running with MPICH2 hydra. >>> >>> Execution command: >>> mpiexec.hydra -genvall -disable-auto-cleanup -f MpiConfigMachines1.txt -launcher=rsh -n 2 mpi_irecv_ft_simple >>> >>> >>> Both hydra's configured with: >>> $ ./configure --prefix=/space/local/mpich-3.0.4/ --enable-error-checking=runtime --enable-g=dbg CFLAGS=-fPIC CXXFLAGS=-fPIC FFLAGS=-fpic --enable-threads=runtime --enable-totalview --enable-static --disable-f77 --disable-fc --no-recursion >>> >>> $ ./configure --prefix=/space/local/mpich2-1.5b2/ --enable-error-checking=runtime --enable-g=dbg CFLAGS=-fPIC CXXFLAGS=-fPIC FFLAGS=-fpic --enable-threads=runtime --enable-totalview --enable-static --disable-f77 --disable-fc >>> >>> Can you advice please? >>> >>> Regards, >>> Anatoly. >>> >>> >>> _______________________________________________ >>> 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] Fault tolerance MPICH2 hydra vs MPICH3.0.4 hydra
by Anatoly G 21 May '14

21 May '14

Thank you, Pavan. I tried MPICH 3.1, it works good with Fault tolerance. One more question: If I execute my simulation (previous mail) with MPICH 3.1 compiled with --with-device=ch3:sock, I see not stable fault tolerance. Sometimes whole system crashes, but sometimes not. If I use default configuration flags (without --with-device=ch3:sock) whole system is stable. Is this expected behavior? Regards, Anatoly. On Mon, May 19, 2014 at 5:03 PM, Balaji, Pavan <balaji(a)anl.gov> wrote: > > Hydra should be compatible between different versions of mpich (or mpich > derivatives). However, there’s always a possibility that there was a bug > in mpich-3.0.4’s hydra that was fixed in mpich-3.1. So we recommend using > the latest version. > > — Pavan > > On May 19, 2014, at 1:15 AM, Anatoly G <anatolyrishon(a)gmail.com> wrote: > > > Hi Wesley. > > Thank you very much for quick response. > > I executed your's code. Master can't finish it's execution. It stalled > on MPI_Wait on iteration 7. > > > > But if I use MPICH2 hydra, Master process will finish executing by > reporting number of times on failure of slave process. > > > > Can you please advice if it's safe to make hybrid system of build with > MPICH3.0.4, but using MPICH2 hydra? > > Or may be any other solution. > > Does MPICH 3.0.4 include all MPICH2 hydra functionality? > > May be my configuration of MPICH 3.0.4 is wrong? > > > > Regards, > > Anatoly. > > > > > > On Sun, May 18, 2014 at 8:40 PM, Wesley Bland <wbland(a)anl.gov> wrote: > > Hi Anatoly, > > > > I think the problem may be the way that you're aborting. MPICH catches > the system abort call and kills the entire application when it's called. > Instead, I suggest using MPI_Abort(MPI_COMM_WORLD, 1); That's what I use in > my tests and it works fine. It also seemed to work for your code when I > tried. I'll attach my modified version of your code. I switched it to C > since I happened to have C++ support disabled on my local install, but that > shouldn't change anything. > > > > Thanks, > > Wesley > > > > > > On Sun, May 18, 2014 at 5:18 AM, Anatoly G <anatolyrishon(a)gmail.com> > wrote: > > Dear MPICH2, > > Can you please help me with understanding Fault Tolerance in MPICH3.0.4 > > I have a simple MPI program: > > Master calls MPI_Irecv + MPI_Wait in loop. > > Single slave: calls MPI_Send x 5 times, then calls abort. > > > > When I execute program with MPICH2 hydra I get multiple times Master > process prints about fail in slave. In MPICH3 hydra I get a single message > about fail of slave and then Master process enters to endless wait for next > Irecv completion. > > In both cases I compiled program with MPICH3.0.4 > > > > In other words, with MPICH2 hydra each Irecv completes (even if slave > died before execution of Irecv) but in MPICH3 hydra not. Causes MPI_Irecv > endless wait. > > > > If I compile same program with MPICH2 and use MPICH2 hydra, I get the > same result as compiling with MPICH3.0.4 and running with MPICH2 hydra. > > > > Execution command: > > mpiexec.hydra -genvall -disable-auto-cleanup -f MpiConfigMachines1.txt > -launcher=rsh -n 2 mpi_irecv_ft_simple > > > > > > Both hydra's configured with: > > $ ./configure --prefix=/space/local/mpich-3.0.4/ > --enable-error-checking=runtime --enable-g=dbg CFLAGS=-fPIC CXXFLAGS=-fPIC > FFLAGS=-fpic --enable-threads=runtime --enable-totalview --enable-static > --disable-f77 --disable-fc --no-recursion > > > > $ ./configure --prefix=/space/local/mpich2-1.5b2/ > --enable-error-checking=runtime --enable-g=dbg CFLAGS=-fPIC CXXFLAGS=-fPIC > FFLAGS=-fpic --enable-threads=runtime --enable-totalview --enable-static > --disable-f77 --disable-fc > > > > Can you advice please? > > > > Regards, > > Anatoly. > > > > > > _______________________________________________ > > 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] invalid rank in mpi_send
by Rob Latham 21 May '14

21 May '14

On 05/21/2014 08:07 AM, Chafik sanaa wrote: > what do you mean by invalid rank ? this is my program : your program runs fine on 2 processes on my linux laptop. Perhaps you mistakenly executed it with one process? if the problem is with MS-MPI, you should try the MS-MPI forum: http://social.microsoft.com/Forums/en-US/home?forum=windowshpcmpi or askmpi(a)microsoft.com ==rob > > #include "mpi.h" > #include <stdlib.h> > #include <stdio.h> > int main(int argc, char *argv[]) { > int rang,iter,iter2; > int nb_tests=9; > int nb_valeurs[9]; > int nb_valeurs_max=7000000; > int etiquette=99; > double *valeurs; > MPI_Status statut; > double temps_debut,temps_fin; > > MPI_Init( &argc, &argv); > > nb_valeurs[0]=0,nb_valeurs[1]=1,nb_valeurs[2]=10,nb_valeurs[3]=100; > nb_valeurs[4]=1000,nb_valeurs[5]=10000,nb_valeurs[6]=100000; > nb_valeurs[7]=1000000,nb_valeurs[8]=7000000; > > MPI_Comm_rank( MPI_COMM_WORLD, &rang); > > valeurs = (double *) malloc(nb_valeurs_max*sizeof(double)); > > for(iter=0; iter<nb_tests; iter++) { > if (rang == 0) { > for (iter2 = 0; iter2<nb_valeurs[iter]; iter2++) > valeurs[iter2] = rand() / (RAND_MAX + 1.); > temps_debut=MPI_Wtime(); > > MPI_Send(valeurs,nb_valeurs[iter],MPI_DOUBLE,1,etiquette,MPI_COMM_WORLD); > MPI_Recv(valeurs,nb_valeurs[iter],MPI_DOUBLE,1,etiquette, > MPI_COMM_WORLD,&statut); > temps_fin=MPI_Wtime(); > if (nb_valeurs[iter] != 0) { > printf("Moi, processus 0, j'ai envoye et recu %8d valeurs" > "(derniere = %4.2f) du processus 1 en %8.6f secondes, soit " > "avec un debit de %7.2f Mo/s.\n", > nb_valeurs[iter], valeurs[nb_valeurs[iter]-1], > temps_fin-temps_debut, > 2.*nb_valeurs[iter]*8/1000000./(temps_fin-temps_debut)); > } else > printf("Moi, processus 0, j'ai envoye et recu %8d valeurs en %8.6f " > "secondes, soit avec un debit de %7.2f Mo/s.\n", > nb_valeurs[iter], temps_fin-temps_debut, > 2.*nb_valeurs[iter]*8/1000000./(temps_fin-temps_debut)); > } else if(rang == 1) { > MPI_Recv(valeurs,nb_valeurs[iter],MPI_DOUBLE,0,etiquette, > MPI_COMM_WORLD,&statut); > > MPI_Send(valeurs,nb_valeurs[iter],MPI_DOUBLE,0,etiquette,MPI_COMM_WORLD); > } > } > > MPI_Finalize(); > return 0; > } > > > > 2014-05-21 15:03 GMT+02:00 Rob Latham <robl(a)mcs.anl.gov > <mailto:[email protected]>>: > > > > On 05/21/2014 04:28 AM, Chafik sanaa wrote: > > Hi, > I use the Microsoft HPC Pack 2008 R2 in the visual studio 2010 > , when i > execute my program i have this error : > fatal error in pmpi_send invalid rank error stack PMPI_Send(150): > MPI_send(buf=... > What is the problem? > > > uh, you passed an invalid rank to MPI_Send. > > ==rob > > thank you > > > _________________________________________________ > To manage subscription options or unsubscribe: > https://lists.mpich.org/__mailman/listinfo/devel > <https://lists.mpich.org/mailman/listinfo/devel> > > > -- > Rob Latham > Mathematics and Computer Science Division > Argonne National Lab, IL USA > _________________________________________________ > To manage subscription options or unsubscribe: > https://lists.mpich.org/__mailman/listinfo/devel > <https://lists.mpich.org/mailman/listinfo/devel> > > > > > _______________________________________________ > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/devel > -- Rob Latham Mathematics and Computer Science Division Argonne National Lab, IL USA

1 0

Re: [mpich-discuss] [mpich-devel] (no subject)
by Wesley Bland 21 May '14

21 May '14

First, this question is now being posted to discuss(a)mpich.org since it's a question about how to use MPICH, not about developing MPICH. Second, don't drop the mailing list from the CC. doing so will not help in getting answers. Third, how are you launching your program? It's probably something like this: mpiexec -n 2 ./my_program On Wed, May 21, 2014 at 8:20 AM, Chafik sanaa <san.chafik(a)gmail.com> wrote: > when i execute my program in MS DOS whith file.exe , i have the error > :fatal error in pmpi_send invalid rank error stack PMPI_Send(150): > MPI_send(buf=... > but when i execute my program whith mpiexec -n 3 file.exe , i have this > error :aborting : failed to launch 'file.exe' on CHAFIK-PC > Error (2) le fichier specifie est introuvable > > > > 2014-05-21 15:13 GMT+02:00 Wesley Bland <wbland(a)anl.gov>: > > How are you launching your program? A first glance at the code looks >> fine. You're probably not using mpiexec correctly. >> >> >> On Wed, May 21, 2014 at 8:07 AM, Chafik sanaa <san.chafik(a)gmail.com>wrote: >> >>> what do you mean by invalid rank ? this is my program : >>> >>> #include "mpi.h" >>> #include <stdlib.h> >>> #include <stdio.h> >>> int main(int argc, char *argv[]) { >>> int rang,iter,iter2; >>> int nb_tests=9; >>> int nb_valeurs[9]; >>> int nb_valeurs_max=7000000; >>> int etiquette=99; >>> double *valeurs; >>> MPI_Status statut; >>> double temps_debut,temps_fin; >>> >>> MPI_Init( &argc, &argv); >>> >>> nb_valeurs[0]=0,nb_valeurs[1]=1,nb_valeurs[2]=10,nb_valeurs[3]=100; >>> nb_valeurs[4]=1000,nb_valeurs[5]=10000,nb_valeurs[6]=100000; >>> nb_valeurs[7]=1000000,nb_valeurs[8]=7000000; >>> >>> MPI_Comm_rank( MPI_COMM_WORLD, &rang); >>> >>> valeurs = (double *) malloc(nb_valeurs_max*sizeof(double)); >>> >>> for(iter=0; iter<nb_tests; iter++) { >>> if (rang == 0) { >>> for (iter2 = 0; iter2<nb_valeurs[iter]; iter2++) >>> valeurs[iter2] = rand() / (RAND_MAX + 1.); >>> temps_debut=MPI_Wtime(); >>> >>> MPI_Send(valeurs,nb_valeurs[iter],MPI_DOUBLE,1,etiquette,MPI_COMM_WORLD); >>> MPI_Recv(valeurs,nb_valeurs[iter],MPI_DOUBLE,1,etiquette, >>> MPI_COMM_WORLD,&statut); >>> temps_fin=MPI_Wtime(); >>> if (nb_valeurs[iter] != 0) { >>> printf("Moi, processus 0, j'ai envoye et recu %8d valeurs" >>> "(derniere = %4.2f) du processus 1 en %8.6f secondes, soit " >>> "avec un debit de %7.2f Mo/s.\n", >>> nb_valeurs[iter], valeurs[nb_valeurs[iter]-1], >>> temps_fin-temps_debut, >>> 2.*nb_valeurs[iter]*8/1000000./(temps_fin-temps_debut)); >>> } else >>> printf("Moi, processus 0, j'ai envoye et recu %8d valeurs en %8.6f " >>> "secondes, soit avec un debit de %7.2f Mo/s.\n", >>> nb_valeurs[iter], temps_fin-temps_debut, >>> 2.*nb_valeurs[iter]*8/1000000./(temps_fin-temps_debut)); >>> } else if(rang == 1) { >>> MPI_Recv(valeurs,nb_valeurs[iter],MPI_DOUBLE,0,etiquette, >>> MPI_COMM_WORLD,&statut); >>> >>> MPI_Send(valeurs,nb_valeurs[iter],MPI_DOUBLE,0,etiquette,MPI_COMM_WORLD); >>> } >>> } >>> >>> MPI_Finalize(); >>> return 0; >>> } >>> >>> >>> >>> 2014-05-21 15:03 GMT+02:00 Rob Latham <robl(a)mcs.anl.gov>: >>> >>> >>>> >>>> On 05/21/2014 04:28 AM, Chafik sanaa wrote: >>>> >>>>> Hi, >>>>> I use the Microsoft HPC Pack 2008 R2 in the visual studio 2010 , when >>>>> i >>>>> execute my program i have this error : >>>>> fatal error in pmpi_send invalid rank error stack PMPI_Send(150): >>>>> MPI_send(buf=... >>>>> What is the problem? >>>>> >>>> >>>> uh, you passed an invalid rank to MPI_Send. >>>> >>>> ==rob >>>> >>>> thank you >>>>> >>>>> >>>>> _______________________________________________ >>>>> To manage subscription options or unsubscribe: >>>>> https://lists.mpich.org/mailman/listinfo/devel >>>>> >>>>> >>>> -- >>>> Rob Latham >>>> Mathematics and Computer Science Division >>>> Argonne National Lab, IL USA >>>> _______________________________________________ >>>> To manage subscription options or unsubscribe: >>>> https://lists.mpich.org/mailman/listinfo/devel >>>> >>> >>> >> >

1 0

Re: [mpich-discuss] [mpich-devel] (no subject)
by Wesley Bland 21 May '14

21 May '14

How are you launching your program? A first glance at the code looks fine. You're probably not using mpiexec correctly. On Wed, May 21, 2014 at 8:07 AM, Chafik sanaa <san.chafik(a)gmail.com> wrote: > what do you mean by invalid rank ? this is my program : > > #include "mpi.h" > #include <stdlib.h> > #include <stdio.h> > int main(int argc, char *argv[]) { > int rang,iter,iter2; > int nb_tests=9; > int nb_valeurs[9]; > int nb_valeurs_max=7000000; > int etiquette=99; > double *valeurs; > MPI_Status statut; > double temps_debut,temps_fin; > > MPI_Init( &argc, &argv); > > nb_valeurs[0]=0,nb_valeurs[1]=1,nb_valeurs[2]=10,nb_valeurs[3]=100; > nb_valeurs[4]=1000,nb_valeurs[5]=10000,nb_valeurs[6]=100000; > nb_valeurs[7]=1000000,nb_valeurs[8]=7000000; > > MPI_Comm_rank( MPI_COMM_WORLD, &rang); > > valeurs = (double *) malloc(nb_valeurs_max*sizeof(double)); > > for(iter=0; iter<nb_tests; iter++) { > if (rang == 0) { > for (iter2 = 0; iter2<nb_valeurs[iter]; iter2++) > valeurs[iter2] = rand() / (RAND_MAX + 1.); > temps_debut=MPI_Wtime(); > > MPI_Send(valeurs,nb_valeurs[iter],MPI_DOUBLE,1,etiquette,MPI_COMM_WORLD); > MPI_Recv(valeurs,nb_valeurs[iter],MPI_DOUBLE,1,etiquette, > MPI_COMM_WORLD,&statut); > temps_fin=MPI_Wtime(); > if (nb_valeurs[iter] != 0) { > printf("Moi, processus 0, j'ai envoye et recu %8d valeurs" > "(derniere = %4.2f) du processus 1 en %8.6f secondes, soit " > "avec un debit de %7.2f Mo/s.\n", > nb_valeurs[iter], valeurs[nb_valeurs[iter]-1], > temps_fin-temps_debut, > 2.*nb_valeurs[iter]*8/1000000./(temps_fin-temps_debut)); > } else > printf("Moi, processus 0, j'ai envoye et recu %8d valeurs en %8.6f " > "secondes, soit avec un debit de %7.2f Mo/s.\n", > nb_valeurs[iter], temps_fin-temps_debut, > 2.*nb_valeurs[iter]*8/1000000./(temps_fin-temps_debut)); > } else if(rang == 1) { > MPI_Recv(valeurs,nb_valeurs[iter],MPI_DOUBLE,0,etiquette, > MPI_COMM_WORLD,&statut); > > MPI_Send(valeurs,nb_valeurs[iter],MPI_DOUBLE,0,etiquette,MPI_COMM_WORLD); > } > } > > MPI_Finalize(); > return 0; > } > > > > 2014-05-21 15:03 GMT+02:00 Rob Latham <robl(a)mcs.anl.gov>: > > >> >> On 05/21/2014 04:28 AM, Chafik sanaa wrote: >> >>> Hi, >>> I use the Microsoft HPC Pack 2008 R2 in the visual studio 2010 , when i >>> execute my program i have this error : >>> fatal error in pmpi_send invalid rank error stack PMPI_Send(150): >>> MPI_send(buf=... >>> What is the problem? >>> >> >> uh, you passed an invalid rank to MPI_Send. >> >> ==rob >> >> thank you >>> >>> >>> _______________________________________________ >>> To manage subscription options or unsubscribe: >>> https://lists.mpich.org/mailman/listinfo/devel >>> >>> >> -- >> Rob Latham >> Mathematics and Computer Science Division >> Argonne National Lab, IL USA >> _______________________________________________ >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/devel >> > >

1 0

Re: [mpich-discuss] [mpich-devel] (no subject)
by Wesley Bland 21 May '14

21 May '14

How are you launching your program? A first glance at the code looks fine. You're probably not using mpiexec correctly. On Wed, May 21, 2014 at 8:07 AM, Chafik sanaa <san.chafik(a)gmail.com> wrote: > what do you mean by invalid rank ? this is my program : > > #include "mpi.h" > #include <stdlib.h> > #include <stdio.h> > int main(int argc, char *argv[]) { > int rang,iter,iter2; > int nb_tests=9; > int nb_valeurs[9]; > int nb_valeurs_max=7000000; > int etiquette=99; > double *valeurs; > MPI_Status statut; > double temps_debut,temps_fin; > > MPI_Init( &argc, &argv); > > nb_valeurs[0]=0,nb_valeurs[1]=1,nb_valeurs[2]=10,nb_valeurs[3]=100; > nb_valeurs[4]=1000,nb_valeurs[5]=10000,nb_valeurs[6]=100000; > nb_valeurs[7]=1000000,nb_valeurs[8]=7000000; > > MPI_Comm_rank( MPI_COMM_WORLD, &rang); > > valeurs = (double *) malloc(nb_valeurs_max*sizeof(double)); > > for(iter=0; iter<nb_tests; iter++) { > if (rang == 0) { > for (iter2 = 0; iter2<nb_valeurs[iter]; iter2++) > valeurs[iter2] = rand() / (RAND_MAX + 1.); > temps_debut=MPI_Wtime(); > > MPI_Send(valeurs,nb_valeurs[iter],MPI_DOUBLE,1,etiquette,MPI_COMM_WORLD); > MPI_Recv(valeurs,nb_valeurs[iter],MPI_DOUBLE,1,etiquette, > MPI_COMM_WORLD,&statut); > temps_fin=MPI_Wtime(); > if (nb_valeurs[iter] != 0) { > printf("Moi, processus 0, j'ai envoye et recu %8d valeurs" > "(derniere = %4.2f) du processus 1 en %8.6f secondes, soit " > "avec un debit de %7.2f Mo/s.\n", > nb_valeurs[iter], valeurs[nb_valeurs[iter]-1], > temps_fin-temps_debut, > 2.*nb_valeurs[iter]*8/1000000./(temps_fin-temps_debut)); > } else > printf("Moi, processus 0, j'ai envoye et recu %8d valeurs en %8.6f " > "secondes, soit avec un debit de %7.2f Mo/s.\n", > nb_valeurs[iter], temps_fin-temps_debut, > 2.*nb_valeurs[iter]*8/1000000./(temps_fin-temps_debut)); > } else if(rang == 1) { > MPI_Recv(valeurs,nb_valeurs[iter],MPI_DOUBLE,0,etiquette, > MPI_COMM_WORLD,&statut); > > MPI_Send(valeurs,nb_valeurs[iter],MPI_DOUBLE,0,etiquette,MPI_COMM_WORLD); > } > } > > MPI_Finalize(); > return 0; > } > > > > 2014-05-21 15:03 GMT+02:00 Rob Latham <robl(a)mcs.anl.gov>: > > >> >> On 05/21/2014 04:28 AM, Chafik sanaa wrote: >> >>> Hi, >>> I use the Microsoft HPC Pack 2008 R2 in the visual studio 2010 , when i >>> execute my program i have this error : >>> fatal error in pmpi_send invalid rank error stack PMPI_Send(150): >>> MPI_send(buf=... >>> What is the problem? >>> >> >> uh, you passed an invalid rank to MPI_Send. >> >> ==rob >> >> thank you >>> >>> >>> _______________________________________________ >>> To manage subscription options or unsubscribe: >>> https://lists.mpich.org/mailman/listinfo/devel >>> >>> >> -- >> Rob Latham >> Mathematics and Computer Science Division >> Argonne National Lab, IL USA >> _______________________________________________ >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/devel >> > >

1 0