[mvapich-discuss] MVPICH2 installcheck
Tong Jun
tongjun at cn.fujitsu.com
Sun Nov 13 22:40:34 EST 2011
Hi MVPICH2 Supporter
I have downloaded, built and installed the "mvapich2-1.7".
When running the "make installcheck" command , the last
two programs are failed with given below info.
Can you tell me how to find the problem of the last two programs?
Thanks,
Tong.
OS: RHEL6 (Virtural Machine--VMware)
=========================================================
Running installation runtest for C collchk program...
*** Test C program with the MPI collective/datatype checking library
..... No.
The failed command is :
/usr/local/bin/mpiexec -n 4 wrong_int_byte
librdmacm: couldn't read ABI version.
librdmacm: assuming: 4
CMA: unable to get RDMA device list
librdmacm: couldn't read ABI version.
librdmacm: assuming: 4
CMA: unable to get RDMA device list
librdmacm: couldn't read ABI version.
librdmacm: assuming: 4
CMA: unable to get RDMA device list
librdmacm: couldn't read ABI version.
librdmacm: assuming: 4
CMA: unable to get RDMA device list
Starting MPI Collective and Datatype Checking!
Backtrace of the callstack at rank 3:
At [0]: ./wrong_int_byte(CollChk_err_han+0x104)[0x41acac]
At [1]: ./wrong_int_byte(CollChk_dtype_scatter+0x177)[0x41bdd2]
At [2]: ./wrong_int_byte(CollChk_dtype_bcast+0x4b)[0x41bc54]
At [3]: ./wrong_int_byte(MPI_Bcast+0x8a)[0x41ab66]
At [4]: ./wrong_int_byte(main+0x79)[0x41aaad]
At [5]: /lib64/libc.so.6(__libc_start_main+0xfd)[0x360b61ec5d]
At [6]: ./wrong_int_byte[0x41a979]
[cli_3]: aborting job:
Fatal error in MPI_Comm_call_errhandler:
Collective Checking: BCAST (Rank 3) --> Inconsistent datatype signatures
detected between rank 3 and rank 0.
Running installation runtest for Fortran collchk program...
*** Test F77 program with the MPI collective/datatype checking library
... No.
The failed command is :
/usr/local/bin/mpiexec -n 4 wrong_reals
librdmacm: couldn't read ABI version.
librdmacm: couldn't read ABI version.
librdmacm: assuming: 4
librdmacm: assuming: 4
CMA: unable to get RDMA device list
CMA: unable to get RDMA device list
librdmacm: couldn't read ABI version.
librdmacm: assuming: 4
CMA: unable to get RDMA device list
librdmacm: couldn't read ABI version.
librdmacm: assuming: 4
CMA: unable to get RDMA device list
Starting MPI Collective and Datatype Checking!
Process 0 of 4 is alive
Process 2 of 4 is alive
Process 3 of 4 is alive
Process 1 of 4 is alive
Backtrace of the callstack at rank 3:
At [0]: ./wrong_reals(CollChk_err_han+0x104)[0x422420]
At [1]: ./wrong_reals(CollChk_dtype_scatter+0x177)[0x423b9e]
At [2]: ./wrong_reals(CollChk_dtype_bcast+0x4b)[0x423a20]
At [3]: ./wrong_reals(MPI_Bcast+0x8a)[0x421e3a]
At [4]: ./wrong_reals(mpi_bcast_+0x4a)[0x4211f6]
At [5]: ./wrong_reals(MAIN__+0x161)[0x41db85]
At [6]: ./wrong_reals(main+0x2a)[0x4d939a]
At [7]: /lib64/libc.so.6(__libc_start_main+0xfd)[0x360b61ec5d]
At [8]: ./wrong_reals[0x41d969]
[cli_3]: aborting job:
Fatal error in MPI_Comm_call_errhandler:
Collective Checking: BCAST (Rank 3) --> Inconsistent datatype signatures
detected between rank 3 and rank 0.
=========================================================
More information about the mvapich-discuss
mailing list