Have you tried the same with MPICH 3.1.4?
Ken
On 07/31/2015 12:52 PM, Grindeanu, Iulian R. wrote:
> hello,
> I am trying to see where is the leak in our code;
> valgrind shows some leaks, like these
>
> ==4262== 1,512 bytes in 3 blocks are still reachable in loss record 6 of 6
> ==4262== at 0x4C2CC70: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
> ==4262== by 0x56A6555: set_collops.part.0 (in /home/iulian/3rdparty/mpich-3.1/install/lib/libmpich.so.12.0.0)
> ==4262== by 0x56A73E2: MPIR_Comm_commit (in /home/iulian/3rdparty/mpich-3.1/install/lib/libmpich.so.12.0.0)
> ==4262== by 0x56DFAD3: MPID_Init (in /home/iulian/3rdparty/mpich-3.1/install/lib/libmpich.so.12.0.0)
> ==4262== by 0x5796878: MPIR_Init_thread (in /home/iulian/3rdparty/mpich-3.1/install/lib/libmpich.so.12.0.0)
> ==4262== by 0x5796288: PMPI_Init (in /home/iulian/3rdparty/mpich-3.1/install/lib/libmpich.so.12.0.0)
> ==4262== by 0x40B20B: main (read_nc.cpp:18)
>
>
> in our code, line 18 is MPI_Init(..)
>
> commenting out portions of our code gets rid of the leaks;
> do you have a suggestion what are we doing wrong, or what is the leak related to? Or could it be a false positive from valgrind?
>
>
> Thanks,
> Iulian
> _______________________________________________
> 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
Hi,
Thanks for the previous post on MPI_Init_thread. Here is the sample program
which failed
#include<cstdio>
#include<cstdlib>
#include<cstring>
#include<iostream>
#include<vector>
#include<pthread.h>
#include<mpi.h>
#include <boost/mpi/environment.hpp>
#include <boost/mpi/communicator.hpp>
#include <boost/serialization/vector.hpp>
#include <boost/mpi.hpp>
using namespace std;
using namespace boost::serialization;
using namespace boost::archive;
namespace mpi = boost::mpi; // using mpi library under boost namespace
mpi::environment env;
mpi::communicator world;
#define MASTER 0
// Aim of this program is to print the Rank of each process and size of
number of MPI Run
int main(int argc, char *argv[])
{
// Declarations and prototypes
int rankID, numprocessors,provided;
// Any serial code goes in here
// Initialize the MPI Environment over here
//MPI_Init(&argc,&argv);
int returnvalue =
MPI_Init_thread(&argc,&argv,MPI_THREAD_MULTIPLE,&provided);
printf("\n VALUE OF PROVIDED IS %d",provided);
printf("\n VALUE OF MPI_THREAD_MULTIPLE IS %d",MPI_THREAD_MULTIPLE);
if(provided < MPI_THREAD_MULTIPLE)
{
printf("\n THREAD LIBRARY DOESN'T HAVE MULTITHREADING
SUPPORT:");
//exit(1);
}
// Terminate the MPI Environment
MPI_Finalize();
}
Error got:
Assertion failed in file
/home/viswa/libraries/mpich-3.1.4/src/include/mpiimplthreadpost.h at line
163: depth > 0 && depth < 10
internal ABORT - process 1
internal ABORT - process 0
I tried to use boost library in my MPI Code. Is MPI_Init_thread
designed to work with Boost libraries?. If not, please suggest an
alternative to
send vectors(STL types) using MPI.
MPI version - 3.1.4
CPU Info:
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 8
On-line CPU(s) list: 0-7
Thread(s) per core: 1
Core(s) per socket: 4
Socket(s): 2
NUMA node(s): 1
Vendor ID: GenuineIntel
CPU family: 6
Model: 44
Stepping: 2
CPU MHz: 3458.000
BogoMIPS: 6916.00
Hypervisor vendor: VMware
Virtualization type: full
L1d cache: 32K
L1i cache: 32K
L2 cache: 256K
L3 cache: 12288K
NUMA node0 CPU(s): 0-7
_______________________________________________
discuss mailing list discuss(a)mpich.org
To manage subscription options or unsubscribe:
https://lists.mpich.org/mailman/listinfo/discuss
