Re: Intel oneAPI with Podman

ugiwgh · ‎06-19-2022

Sorry, I cannot reply any comments. So I post a new one.

I use macOS Big Sur, version 11.1(20C69), and Google Chrome version 100.0.4896.75. I change many ways, still cannot reply an comments. Such as, safari(on macOS), Microsoft Edge.

$ mpirun --version

Intel(R) MPI Library for Linux* OS, Version 2021.5 Build 20211102 (id: 9279b7d62)

$ podman --version

podman version 4.0.2

$ podman version

Client: Podman Engine

Version: 4.0.2

API Version: 4.0.2

Go Version: go1.16.13

Git Commit: c99f9f1b6960b98158b7f5fc4e6b5ac1a10b3542

Built: Wed Mar 23 18:33:41 2022

OS/Arch: linux/amd64

$ cat /etc/centos-release

CentOS Linux release 7.9.2009 (Core)

$ uname -a

Linux ln131.para.bscc 3.10.0-1160.25.1.el7.x86_64 #1 SMP Wed Apr 28 21:49:45 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux

The code is intel/oneapi/mpi/2021.5.1/test/test.c.

#include "mpi.h"
#include <stdio.h>
#include <string.h>

int main(int argc, char *argv[])
{
    int i, rank, size, namelen;
    char name[MPI_MAX_PROCESSOR_NAME];
    MPI_Status stat;

    MPI_Init(&argc, &argv);

    MPI_Comm_size(MPI_COMM_WORLD, &size);
    MPI_Comm_rank(MPI_COMM_WORLD, &rank);
    MPI_Get_processor_name(name, &namelen);

    if (rank == 0) {
        printf("Hello world: rank %d of %d running on %s\n", rank, size, name);

        for (i = 1; i < size; i++) {
            MPI_Recv(&rank, 1, MPI_INT, i, 1, MPI_COMM_WORLD, &stat);
            MPI_Recv(&size, 1, MPI_INT, i, 1, MPI_COMM_WORLD, &stat);
            MPI_Recv(&namelen, 1, MPI_INT, i, 1, MPI_COMM_WORLD, &stat);
            MPI_Recv(name, namelen + 1, MPI_CHAR, i, 1, MPI_COMM_WORLD, &stat);
            printf("Hello world: rank %d of %d running on %s\n", rank, size, name);
        }
    } else {
        MPI_Send(&rank, 1, MPI_INT, 0, 1, MPI_COMM_WORLD);
        MPI_Send(&size, 1, MPI_INT, 0, 1, MPI_COMM_WORLD);
        MPI_Send(&namelen, 1, MPI_INT, 0, 1, MPI_COMM_WORLD);
        MPI_Send(name, namelen + 1, MPI_CHAR, 0, 1, MPI_COMM_WORLD);
    }

    MPI_Finalize();
    return (0);
}

I use the following command. The envfile is mpi enviroment in container.

mpirun -np 1 podman run --env-host --env-file envfile --userns=keep-id --network=host --pid=host --ipc=host -w /exports -v .:/exports:z dptech/vasp:5.4.4 /exports/mpitest

The following is envfile.


MKLROOT=/opt/intel/oneapi/mkl/2022.0.2
MANPATH=/opt/intel/oneapi/mpi/2021.5.1/man:/opt/intel/oneapi/debugger/2021.5.0/documentation/man:/opt/intel/oneapi/compiler/2022.0.2/documentation/en/man/common::
LIBRARY_PATH=/opt/intel/oneapi/tbb/2021.5.1/env/../lib/intel64/gcc4.8:/opt/intel/oneapi/mpi/2021.5.1//libfabric/lib:/opt/intel/oneapi/mpi/2021.5.1//lib/release:/opt/intel/oneapi/mpi/2021.5.1//lib:/opt/intel/oneapi/mkl/2022.0.2/lib/intel64:/opt/intel/oneapi/compiler/2022.0.2/linux/compiler/lib/intel64_lin:/opt/intel/oneapi/compiler/2022.0.2/linux/lib
OCL_ICD_FILENAMES=libintelocl_emu.so:libalteracl.so:/opt/intel/oneapi/compiler/2022.0.2/linux/lib/x64/libintelocl.so
LD_LIBRARY_PATH=/home/rsync/.local/container/lib64:/home/rsync/.local/container/lib:/opt/intel/oneapi/tbb/2021.5.1/env/../lib/intel64/gcc4.8:/opt/intel/oneapi/mpi/2021.5.1//libfabric/lib:/opt/intel/oneapi/mpi/2021.5.1//lib/release:/opt/intel/oneapi/mpi/2021.5.1//lib:/opt/intel/oneapi/mkl/2022.0.2/lib/intel64:/opt/intel/oneapi/debugger/2021.5.0/gdb/intel64/lib:/opt/intel/oneapi/debugger/2021.5.0/libipt/intel64/lib:/opt/intel/oneapi/debugger/2021.5.0/dep/lib:/opt/intel/oneapi/compiler/2022.0.2/linux/lib:/opt/intel/oneapi/compiler/2022.0.2/linux/lib/x64:/opt/intel/oneapi/compiler/2022.0.2/linux/lib/oclfpga/host/linux64/lib:/opt/intel/oneapi/compiler/2022.0.2/linux/compiler/lib/intel64_lin
FPGA_VARS_DIR=/opt/intel/oneapi/compiler/2022.0.2/linux/lib/oclfpga
FI_PROVIDER_PATH=/opt/intel/oneapi/mpi/2021.5.1//libfabric/lib/prov:/usr/lib64/libfabric
CMPLR_ROOT=/opt/intel/oneapi/compiler/2022.0.2
CPATH=/opt/intel/oneapi/tbb/2021.5.1/env/../include:/opt/intel/oneapi/mpi/2021.5.1//include:/opt/intel/oneapi/mkl/2022.0.2/include:/opt/intel/oneapi/dev-utilities/2021.5.2/include
INTELFPGAOCLSDKROOT=/opt/intel/oneapi/compiler/2022.0.2/linux/lib/oclfpga
NLSPATH=/opt/intel/oneapi/mkl/2022.0.2/lib/intel64/locale/%l_%t/%N:/opt/intel/oneapi/compiler/2022.0.2/linux/compiler/lib/intel64_lin/locale/%l_%t/%N
PATH=/home/rsync/.local/container/bin:/home/rsync/.local/container/sbin:/opt/intel/oneapi/mpi/2021.5.1//libfabric/bin:/opt/intel/oneapi/mpi/2021.5.1//bin:/opt/intel/oneapi/mkl/2022.0.2/bin/intel64:/opt/intel/oneapi/dev-utilities/2021.5.2/bin:/opt/intel/oneapi/debugger/2021.5.0/gdb/intel64/bin:/opt/intel/oneapi/compiler/2022.0.2/linux/lib/oclfpga/bin:/opt/intel/oneapi/compiler/2022.0.2/linux/bin/intel64:/opt/intel/oneapi/compiler/2022.0.2/linux/bin:/home/rsync/.local/bin:/public1/clurm/bin:/usr/lib64/qt-3.3/bin:/usr/local/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/opt/ibutils/bin:/home/rsync/.local/bin:/home/rsync/bin:/opt/vasp.5.4.4/bin
TBBROOT=/opt/intel/oneapi/tbb/2021.5.1/env/..
INTEL_PYTHONHOME=/opt/intel/oneapi/debugger/2021.5.0/dep
CLASSPATH=/opt/intel/oneapi/mpi/2021.5.1//lib/mpi.jar
PKG_CONFIG_PATH=/opt/intel/oneapi/tbb/2021.5.1/env/../lib/pkgconfig:/opt/intel/oneapi/mpi/2021.5.1/lib/pkgconfig:/opt/intel/oneapi/mkl/2022.0.2/lib/pkgconfig:/opt/intel/oneapi/compiler/2022.0.2/lib/pkgconfig
GDB_INFO=/opt/intel/oneapi/debugger/2021.5.0/documentation/info/
INFOPATH=/opt/intel/oneapi/debugger/2021.5.0/gdb/intel64/lib
CMAKE_PREFIX_PATH=/opt/intel/oneapi/tbb/2021.5.1/env/..:/opt/intel/oneapi/compiler/2022.0.2/linux/IntelDPCPP
ONEAPI_ROOT=/opt/intel/oneapi
I_MPI_ROOT=/opt/intel/oneapi/mpi/2021.5.1
_=/opt/intel/oneapi/mpi/2021.5.1/bin/mpiexec.hydra

HemanthCH_Intel · ‎06-20-2022

Hi,

Thanks for posting in Intel communities.

We tried to run the mpi program using podman, but we are facing the issue as permission denied from the docker registry as shown below screenshot.

Is it a private registry? If yes, is it possible to give access to the registry?

Thanks & Regards,

Hemanth

HemanthCH_Intel · ‎06-27-2022

Hi,

As you are facing issues to respond to the existing post in the Intel communities. So, we have contacted you privately, please check your inbox and respond back to us.

Thanks & Regards,

Hemanth.

HemanthCH_Intel · ‎07-03-2022

Hi,

As we are communicating internally through emails, Since this is a duplicate thread of https://community.intel.com/t5/Intel-oneAPI-HPC-Toolkit/Re-Intel-oneAPI-with-Podman/m-p/1394023#M9611, we will no longer monitor this thread. We will continue addressing this issue in the other thread

Thanks & Regards,

Hemanth