Dear Axel and lammps users,
I was running TAD without any problems and this week I installed on two lammps-11Aug17 computers with the user-omp package. On those two computers, things are strange and I can not even run “in.tad” in the examples folder. The program does not stop, but nothing really happens. Is there any reason for this?
Fernanda S Teixeira
Dear Axel and lammps users,
I was running TAD without any problems and this week I installed on two
lammps-11Aug17 computers with the user-omp package. On those two computers,
things are strange and I can not even run "in.tad" in the examples folder.
The program does not stop, but nothing really happens. Is there any reason
for this?
does this also happen, after you uninstall the USER-OMP package?
axel.
Hi Axel,
No, this does not happen.
Well, I think I do not need OMP package at all, but I would like to understand what the problem is.
Fernanda.
Hi Axel,
No, this does not happen.
Well, I think I do not need OMP package at all, but I would like to
understand what the problem is.
then you have to start debugging. i would like to know that, too.
axel.
i just ran a test with the current LAMMPS development version without USER-OMP installed (and without OpenMP enabled) and it fails the TAD input example with MPI errors, too. so it looks to me, that USER-OMP is not to blame here, but that there are some stray MPI messages causing trouble.
axel.
]$ mpirun -np 3 ~/compile/lammps-icms/src/lmp_mpi -p 3x1 -in in.tad
LAMMPS (22 Sep 2017)
Running on 3 partitions of processors
Setting up TAD …
Step CPU N M Status Barrier Margin t_lo delt_lo
1000 0.000 0 0 E 0.000 0.000 0.000 0.000
Fatal error in PMPI_Allgather: Message truncated, error stack:
PMPI_Allgather(1000)…: MPI_Allgather(sbuf=0x7ffe97b3bd10, scount=1, MPI_DOUBLE, rbuf=0x3670450, rcount=1, MPI_DOUBLE, comm=0x84000005) failed
MPIR_Allgather_impl(840)…:
MPIR_Allgather(799)…:
MPIR_Allgather_intra(534)…:
MPIDI_CH3U_Receive_data_found(131): Message from rank 1 and tag 7 truncated; 32 bytes received but buffer size is 8
MPIR_Allgather_intra(557)…:
MPIDI_CH3U_Receive_data_found(131): Message from rank 2 and tag 7 truncated; 32 bytes received but buffer size is 8
Fatal error in PMPI_Allgather: Other MPI error, error stack:
PMPI_Allgather(1000)…: MPI_Allgather(sbuf=0x7ffec97d9240, scount=4, MPI_DOUBLE, rbuf=0x398bb00, rcount=4, MPI_DOUBLE, comm=0x84000003) failed
MPIR_Allgather_impl(840).:
MPIR_Allgather(799)…:
MPIR_Allgather_intra(634): Failure during collective
Hi Axel,
In my computer, the process gets stuck in the middle of TAD, but I got no error messages.
Fernanda