[OMPI users] Query:File locking failed in ADIOI_Set_lock
Hello, I am getting this following error when I run my application on more than 1 nodes. File locking failed in ADIOI_Set_lock. If the file system is NFS, you need to use NFS version 3, ensure that the lockd daemon is running on all the machines, and mount the directory with the 'noac' option (no attribute caching). I have NFS version 4, and we have mounted the dir with noac option. Will it work only with NFS 3 version only? Any help will be appreciated. Thanks Raj Send instant messages to your online friends http://uk.messenger.yahoo.com
Re: [OMPI users] Problem Installing OpenMPI with Lahey 64-bit Fortran compiler
This post was accidentally sent to the list owner rather than to the entire list. Re-sending to the list... On May 1, 2008, at 11:28 AM, cfdm...@aim.com wrote: I am encountering a difficulty when installing OpenMPI with the Lahey 64-bit (LF64 V8.0) Fortran Compiler for use on a dual quad- core machine. I am currently trying to install this particular Lahey compiler with openmpi-1.2.6 on a 64-bit intel machine running Red Hat Linux Enterprise 5. In the past, I have not had any difficulty installing 32-bit versions of the Lahey compiler with the current or previous versions of OpenMPI, so this issue appears to be specific to the 64-bit version of the Lahey compiler. I have attached the install script that I used to install OpenMPI. I have also attached the output stream that results when I execute this install script. The error message that it is showing is: /usr/bin/ld: .libs/mpi_wtick_f90.o: relocation R_X86_64_32S against `a local symbol' can not be used when making a shared object; recompile with -fPIC We don't test with the Laheay compiler, so I'm not familiar with the exact problem here -- perhaps try adding -fPIC to the FCFLAGS when running configure? ./configure FCFLAGS=-fPIC ... -- Jeff Squyres Cisco Systems config.log.bz2 Description: Binary data configure.log.bz2 Description: Binary data make.log.bz2 Description: Binary data
Re: [OMPI users] Install BLACS and ScaLAPACK on Leopard
FWIW, I'm not a fortran expert, but if you built your Fortran libraries with g77 and then tried to link against them with gfortran, you might run into problems. My advice would be to use a single fortran compiler for building everything: Open MPI, your libraries, your apps. I prefer gfortran because it's more modern, but I have not done any performance evaluations of gfortran vs. g77 -- I have heard [unverified] anecdotes that gfortran is "slower" than g77 -- google around and see what the recent buzz is. FWIW: I tend to use the gnu suite from http://hpc.sourceforge.net/ -- it contains pre-built gcc/g++/gfortran binaries and libraries for Leopard. On May 5, 2008, at 2:59 PM, Linwei Wang wrote: Dear Reeder, It does not work. I do think they are from the fortran programs I'm using (they are files included from the BLACS installation package, not written by my own. The thing is last time when I was using g77, it caused no problem... thanks for your help. Linwei. On May 5, 2008, at 2:33 PM, Doug Reeder wrote: _s_wsle, _e_wsle ___ users mailing list us...@open-mpi.org http://www.open-mpi.org/mailman/listinfo.cgi/users -- Jeff Squyres Cisco Systems
Re: [OMPI users] Install BLACS and ScaLAPACK on Leopard
Thanks, Jeff, I finally choose to use the ScaLAPACK installer provided on the website~ with gcc/gfortran and openmpi...it is installed successfully but I haven't tested the applications.. Thanks for your help. Linwei On May 6, 2008, at 10:09 AM, Jeff Squyres wrote: FWIW, I'm not a fortran expert, but if you built your Fortran libraries with g77 and then tried to link against them with gfortran, you might run into problems. My advice would be to use a single fortran compiler for building everything: Open MPI, your libraries, your apps. I prefer gfortran because it's more modern, but I have not done any performance evaluations of gfortran vs. g77 -- I have heard [unverified] anecdotes that gfortran is "slower" than g77 -- google around and see what the recent buzz is. FWIW: I tend to use the gnu suite from http://hpc.sourceforge.net/ -- it contains pre-built gcc/g++/gfortran binaries and libraries for Leopard. On May 5, 2008, at 2:59 PM, Linwei Wang wrote: Dear Reeder, It does not work. I do think they are from the fortran programs I'm using (they are files included from the BLACS installation package, not written by my own. The thing is last time when I was using g77, it caused no problem... thanks for your help. Linwei. On May 5, 2008, at 2:33 PM, Doug Reeder wrote: _s_wsle, _e_wsle ___ users mailing list us...@open-mpi.org http://www.open-mpi.org/mailman/listinfo.cgi/users -- Jeff Squyres Cisco Systems ___ users mailing list us...@open-mpi.org http://www.open-mpi.org/mailman/listinfo.cgi/users
Re: [OMPI users] Install BLACS and ScaLAPACK on Leopard
Points to clarify if I may, having gone through this relatively recently: g77 and gfortran are NOT one and the same. gfortran from sourceforge works well, but it is based on gnu gcc 4.3 and not on the gnu gcc 4.0.1 that comes with Leopard. Your best bet is to download the ENTIRE gcc package from sourceforge and install it into /usr/local. This includes gcc, g++, and gfortran. Then you will need to do a number of things to actually get a reliable set of packages all compiled from the same version of gcc 4.3. Why? Because 4.3 seems to be notoriously faster. AND, I had a lot of problems integrating the 4.0.1 libs with the 4.3 libs without errors 1. download CLAPACK-3.1.1 from netlib And compile 2. Download ATLAS-1.8 from dourceforge (netlib is a little behind here) and configure it with the --with-netlib-lapack=your just compiled lapack from CLAPACK 3. Download OpenMPI 1.2.6 and install it also so that openMPI will have the fortran not installed with Leopard. 4. NOW you can compile BLACS and ScaLAPACK In all of this you will need to do a couple of additional things like set the env's setenv LDFLAGS "-L/usr/local/lib/x86_64" setenv DYLD_LIBRARY_PATH "your openmpi path" setenv LD_LIBRARY_PATH "your openmpi path" Do all this right and make sure you compile with the -m64 -mtune=core2 flags and you will be golden. So what will you have--- A new cblas, atlas, lapack, openmpi, fortran, c, c++, blacs, and scalapack. All on the same version of gnu c. Alternatively you can buy and use the intel compiler. It is significantly faster than gfortran, but it has a host of other problems associated with it. But if you follow the outline above, you will be left with the best that's available. I have lots more info on this, but time is short. FINALLY, and this is important, DO NOT FORGET ABOUT THE small STACK size on Mac's when using gfortran. It's so small that it's useless for large parallel jobs. On May 6, 2008, at 10:09 AM, Jeff Squyres wrote: FWIW, I'm not a fortran expert, but if you built your Fortran libraries with g77 and then tried to link against them with gfortran, you might run into problems. My advice would be to use a single fortran compiler for building everything: Open MPI, your libraries, your apps. I prefer gfortran because it's more modern, but I have not done any performance evaluations of gfortran vs. g77 -- I have heard [unverified] anecdotes that gfortran is "slower" than g77 -- google around and see what the recent buzz is. FWIW: I tend to use the gnu suite from http://hpc.sourceforge.net/ -- it contains pre-built gcc/g++/gfortran binaries and libraries for Leopard. On May 5, 2008, at 2:59 PM, Linwei Wang wrote: Dear Reeder, It does not work. I do think they are from the fortran programs I'm using (they are files included from the BLACS installation package, not written by my own. The thing is last time when I was using g77, it caused no problem... thanks for your help. Linwei. On May 5, 2008, at 2:33 PM, Doug Reeder wrote: _s_wsle, _e_wsle ___ users mailing list us...@open-mpi.org http://www.open-mpi.org/mailman/listinfo.cgi/users -- Jeff Squyres Cisco Systems ___ users mailing list us...@open-mpi.org http://www.open-mpi.org/mailman/listinfo.cgi/users
[OMPI users] Problem with AlltoAll routine
Hi, i tried to run SkaMPI 5.0.4 benchmark on IBM-BladeCenterLS21 system with 256 processors over Infiniband 5 Gb/s, but test has stopped on "AlltoAll-length" routine, with count=2048 for some reason. I have launched test with: --mca btl_openib_eager_limit 1024 Same tests with 128 processor or less, have finished successful. Different values of eager limit don't solve the problem. Version of OpenMPI involved is 1.2.5. There's someone with this kind of problem over Infiniband? Thanks in advance. -- Gabriele Fatigati CINECA Systems & Tecnologies Department Supercomputing Group Via Magnanelli 6/3, Casalecchio di Reno (BO) Italy www.cineca.itTel:39 051 6171722 g.fatig...@cineca.it
Re: [OMPI users] MPI_Lookup_name error
Correct, I am using 1.2.6 and not running a persistent deamon (thank you for
the link Pak by the way).
However, in the Client/Server test thread I'm showing a complete example
where I tried to run both applications through the same mpirun command and
still having an internal error:
http://www.open-mpi.org/community/lists/users/2008/04/5537.php
On Mon, May 5, 2008 at 10:18 AM, Ralph Castain wrote:
> I assume you are using some variant of OMPI 1.2?
>
> When last I checked, which admittedly was quite a while ago, this worked
> on
> the 1.2.x series. However, I note something here that may be a problem. In
> the 1.2.x series, we do not have a global publish/lookup service - the
> application doing the publish must be launched by the same mpirun as the
> application doing the lookup.
>
> The code below only does the lookup, and appears to be asking that you
> provide some server name. I assume you are somehow looking up the name of
> the mpirun that launched the application that did the publish, and hoping
> the two will cross-connect? Unfortunately, I don't believe the 1.2.x code
> is
> smart enough to figure out how to do that.
>
> This is cleaned up in the upcoming 1.3 release and should work much
> smoother.
>
> Ralph
>
>
>
> On 4/27/08 6:58 PM, "Alberto Giannetti"
> wrote:
>
> > I am having error using MPI_Lookup_name. The same program works fine
> > when using MPICH:
> >
> >
> > /usr/local/bin/mpiexec -np 2 ./client myfriend
> > Processor 0 (662, Sender) initialized
> > Processor 0 looking for service myfriend-0
> > Processor 1 (664, Sender) initialized
> > Processor 1 looking for service myfriend-1
> > [local:00662] *** An error occurred in MPI_Lookup_name
> > [local:00662] *** on communicator MPI_COMM_WORLD
> > [local:00662] *** MPI_ERR_NAME: invalid name argument
> > [local:00662] *** MPI_ERRORS_ARE_FATAL (goodbye)
> > [local:00664] *** An error occurred in MPI_Lookup_name
> > [local:00664] *** on communicator MPI_COMM_WORLD
> > [local:00664] *** MPI_ERR_NAME: invalid name argument
> > [local:00664] *** MPI_ERRORS_ARE_FATAL (goodbye)
> >
> >
> > int main(int argc, char* argv[])
> > {
> >int rank, i;
> >float data[100];
> >char cdata[64];
> >char myport[MPI_MAX_PORT_NAME];
> >char myservice[64];
> >MPI_Comm intercomm;
> >MPI_Status status;
> >int intercomm_size;
> >
> >MPI_Init(&argc, &argv);
> >MPI_Comm_rank(MPI_COMM_WORLD, &rank);
> >printf("Processor %d (%d, Sender) initialized\n", rank, getpid());
> >
> >if( argc < 2 ) {
> > fprintf(stderr, "Require server name\n");
> > MPI_Finalize();
> > exit(-1);
> >}
> >
> >for( i = 0; i < 100; i++ )
> > data[i] = i;
> >
> >sprintf(myservice, "%s-%d", argv[1], rank);
> >printf("Processor %d looking for service %s\n", rank, myservice);
> >MPI_Lookup_name(myservice, MPI_INFO_NULL, myport);
> >printf("Processor %d found port %s looking for service %s\n",
> > rank, myport, myservice);
> >
> >while( 1 ) {
> > printf("Processor %d connecting to '%s'\n", rank, myport);
> > if( MPI_Comm_connect(myport, MPI_INFO_NULL, 0, MPI_COMM_SELF,
> > &intercomm) == MPI_SUCCESS )
> >break;
> > sleep(1);
> >}
> >printf("Processor %d connected\n", rank);
> >
> >MPI_Comm_remote_size(intercomm, &intercomm_size);
> >printf("Processor %d remote comm size is %d\n", rank,
> > intercomm_size);
> >
> >printf("Processor %d sending data through intercomm to rank 0...
> > \n", rank);
> >MPI_Send(data, 100, MPI_FLOAT, 0, rank, intercomm);
> >printf("Processor %d data sent!\n", rank);
> >MPI_Recv(cdata, 12, MPI_CHAR, MPI_ANY_SOURCE, MPI_ANY_TAG,
> > intercomm, &status);
> >printf("Processor %d received string data '%s' from rank %d, tag %d
> > \n", rank, cdata, status.MPI_SOURCE, status.MPI_TAG);
> >
> >sleep(5);
> >
> >printf("Processor %d disconnecting communicator\n", rank);
> >MPI_Comm_disconnect(&intercomm);
> >printf("Processor %d finalizing\n", rank);
> >
> >MPI_Finalize();
> >printf("Processor %d Goodbye!\n", rank);
> > }
> >
> > ___
> > users mailing list
> > us...@open-mpi.org
> > http://www.open-mpi.org/mailman/listinfo.cgi/users
>
>
> ___
> users mailing list
> us...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/users
>
Re: [OMPI users] Problem with AlltoAll routine
Hello Gabriele, To help track down this problem, could I ask you to take a look at the Open MPI "Getting Help" page? http://www.open-mpi.org/community/help/ In particular, if you could collect and send the information requested on that page to the list, it will help us to better understand your configuration and how others might reproduce the problem. Thanks & Regards, --Brad Brad Benton IBM On Tue, May 6, 2008 at 10:35 AM, Gabriele FATIGATI wrote: > Hi, > i tried to run SkaMPI 5.0.4 benchmark on IBM-BladeCenterLS21 system with > 256 processors over Infiniband 5 Gb/s, but test has stopped on > "AlltoAll-length" routine, with count=2048 for some reason. > > I have launched test with: > --mca btl_openib_eager_limit 1024 > > Same tests with 128 processor or less, have finished successful. > > Different values of eager limit don't solve the problem. Version of > OpenMPI involved is 1.2.5. There's someone with this kind of problem over > Infiniband? > Thanks in advance. > -- > Gabriele Fatigati > > CINECA Systems & Tecnologies Department > > Supercomputing Group > > Via Magnanelli 6/3, Casalecchio di Reno (BO) Italy > > www.cineca.itTel:39 051 6171722 > > g.fatig...@cineca.it > ___ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users >
