ICODE-ADC/lammps
4
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

git-svn-id: svn://svn.icms.temple.edu/lammps-ro/trunk@7333 f3b2605a-c512-4ea7-a41b-209d697bcdaa

Browse Source
This commit is contained in:
sjplimp
2011-12-09 16:46:31 +00:00
parent 7060c62c8d
commit 5baa07d3af
8 changed files with 63 additions and 574 deletions
Download Patch File Download Diff File Expand all files Collapse all files

32
src/comm.cpp
View File

@ -1201,8 +1201,9 @@ void Comm::reverse_comm_dump(Dump *dump)
area = surface area of each of 3 faces of simulation box divided by sx,sy,sz
for triclinic, area = cross product of 2 edge vectors stored in h matrix
valid assignment will be factorization of numprocs = Px by Py by Pz
user_factors = if non-zero, factor is specified by user
user_factors = if non-zero, factors are specified by user
sx,sy,sz = scale box xyz dimension vy dividing by sx,sy,sz
other = 1 to enforce compatability with other partition's layout
return factors = # of procs assigned to each dimension
return 1 if successully factor, 0 if not
------------------------------------------------------------------------- */
@ -1553,6 +1554,16 @@ void Comm::set_processors(int narg, char **arg)
} else error->all(FLERR,"Illegal processors command");
}
// error check
if (numa_nodes) {
if (layoutflag != CART)
error->all(FLERR,"Can only use processors numa "
"with processors grid cart");
if (send_to_partition >= 0 || recv_from_partition >= 0)
error->one(FLERR,"Cannot use processors numa with processors part");
}
}
/* ----------------------------------------------------------------------
@ -1587,15 +1598,15 @@ int Comm::numa_set_proc_grid()
int procs_per_node = name_map.begin()->second;
int procs_per_numa = procs_per_node / numa_nodes;
// use regular mapping if any condition met
// use non-numa mapping if any condition met
if (procs_per_numa < 4 || // less than 4 procs per numa node
procs_per_node % numa_nodes != 0 || // reserve usage for numa_node != 1
procs_per_node % numa_nodes != 0 || // no-op since numa_nodes = 1 for now
nprocs % procs_per_numa != 0 || // total procs not a multiple of node
nprocs <= procs_per_numa || // only 1 node used
user_procgrid[0] > 1 || // user specified grid dimension
user_procgrid[1] > 1 || // that is greater than 1
user_procgrid[2] > 1) { // in any dimension
user_procgrid[0] > 1 || // user specified grid dimension < 1
user_procgrid[1] > 1 || // in any dimension
user_procgrid[2] > 1) {
if (me == 0) {
if (screen) fprintf(screen," 1 by 1 by 1 Node grid\n");
if (logfile) fprintf(logfile," 1 by 1 by 1 Node grid\n");
@ -1617,17 +1628,13 @@ int Comm::numa_set_proc_grid()
// get an initial factorization for each numa node,
// if the user has not set the number of processors
// can fail (on one partition) if constrained by other_partition_style
int numagrid[3];
int flag = procs2box(procs_per_numa,user_numagrid,numagrid,
1,1,1,other_partition_style);
if (!flag) error->all(FLERR,"Could not layout grid of processors",1);
procs2box(procs_per_numa,user_numagrid,numagrid,1,1,1,0);
if (numagrid[0] * numagrid[1] * numagrid[2] != procs_per_numa)
error->all(FLERR,"Bad node grid of processors");
// get a factorization for the grid of numa nodes
// should never fail
int node_count = nprocs / procs_per_numa;
procs2box(node_count,user_procgrid,procgrid,
@ -1637,7 +1644,6 @@ int Comm::numa_set_proc_grid()
// repeat the numa node factorization using the subdomain sizes
// this will refine the factorization if the user specified the node layout
// should never fail
procs2box(procs_per_numa,user_numagrid,numagrid,
procgrid[0],procgrid[1],procgrid[2],0);
@ -1675,7 +1681,7 @@ int Comm::numa_set_proc_grid()
MPI_Comm_split(world,numa_rank,0,&numa_leaders);
// use the MPI Cartesian routines to map the nodes to the grid
// could implement layoutflag as in non-NUMA case
// could implement layoutflag as in non-NUMA case?
int reorder = 0;
int periods[3];