This file is indexed.

/usr/lib/python2.7/dist-packages/PySPH-1.0a4.dev0-py2.7-linux-x86_64.egg/pysph/parallel/tests/lb_exchange.py is in python-pysph 0~20160514.git91867dc-4build1.

This file is owned by root:root, with mode 0o644.

The actual contents of the file can be viewed below.

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
"""Test the ParticleArrayExchange object with the following data

25 particles are created on a rectangular grid with the following
processor assignment:

2---- 3---- 3---- 3---- 3
|     |     |     |     |
1---- 1---- 1---- 2---- 2
|     |     |     |     |
0---- 0---- 1---- 1---- 1
|     |     |     |     |
0---- 0---- 0---- 0---- 0
|     |     |     |     |
0---- 0---- 0---- 0---- 0


We assume that after a load balancing step, we have the following
distribution:

1---- 1---- 1---- 3---- 3
|     |     |     |     |
1---- 1---- 1---- 3---- 3
|     |     |     |     |
0---- 0---- 0---- 3---- 3
|     |     |     |     |
0---- 0---- 2---- 2---- 2
|     |     |     |     |
0---- 0---- 2---- 2---- 2

We create a ParticleArray to represent the initial distribution and
use ParticleArrayExchange to move the data by manually setting the
particle import/export lists. We require that the test be run with 4
processors.

"""
import mpi4py.MPI as mpi
import numpy as np

from pysph.parallel.parallel_manager import ParticleArrayExchange
from pysph.base.utils import get_particle_array_wcsph

comm = mpi.COMM_WORLD
rank = comm.Get_rank()
size = comm.Get_size()

if size != 4:
    if rank == 0:
        raise RuntimeError("Run this test with 4 processors")

# create the initial distribution
if rank == 0:
    numPoints = 12
    x = np.array( [0.0, 1.0, 2.0, 3.0, 4.0,
                   0.0, 1.0, 2.0, 3.0, 4.0,
                   0.0, 1.0] )

    y = np.array( [0.0, 0.0, 0.0, 0.0, 0.0,
                   1.0, 1.0, 1.0, 1.0, 1.0,
                   2.0, 2.0] )

    gid = np.array( [0, 1, 2, 3, 4, 5,
                     6, 7, 8, 9, 10,
                     11], dtype=np.uint32 )

    exportLocalids = np.array( [2, 3, 4, 7, 8, 9], dtype=np.uint32 )
    exportProcs = np.array( [2, 2, 2, 2, 2, 2], dtype=np.int32 )
    numExport = 6

    parts = np.ones(shape=numPoints, dtype=np.int32) * rank

if rank == 1:
    numPoints = 6
    x = np.array( [2.0, 3.0, 4.0,
                   0.0, 1.0, 2.0] )

    y = np.array( [2.0, 2.0, 2.0,
                   3.0, 3.0, 3.0] )

    gid = np.array( [12, 13, 14, 15,
                     16, 17], dtype=np.uint32 )

    exportLocalids = np.array( [0, 1, 2], dtype=np.uint32 )
    exportProcs = np.array( [0, 3, 3], dtype=np.int32 )
    numExport = 3

    parts = np.ones(shape=numPoints, dtype=np.int32) * rank

if rank == 2:
    numPoints = 3
    x = np.array( [4.0, 5.0, 0.0] )
    y = np.array( [3.0, 3.0, 4.0] )
    gid = np.array( [18, 19, 20], dtype=np.uint32 )

    exportLocalids = np.array( [0, 1, 2], dtype=np.uint32 )
    exportProcs = np.array( [3, 3, 1], dtype=np.int32 )
    numExport = 3

    parts = np.ones(shape=numPoints, dtype=np.int32) * rank

if rank == 3:
    numPoints = 4
    x = np.array( [1.0, 2.0, 3.0, 4.0] )
    y = np.array( [4.0, 4.0, 4.0, 4.0] )
    gid = np.array( [21, 22, 23, 24], dtype=np.uint32 )

    exportLocalids = np.array( [0,1], dtype=np.uint32 )
    exportProcs = np.array( [1, 1], dtype=np.int32 )
    numExport = 2

    parts = np.ones(shape=numPoints, dtype=np.int32) * rank

# Gather the Global data on root
X = np.zeros(shape=25, dtype=np.float64)
Y = np.zeros(shape=25, dtype=np.float64)
GID = np.zeros(shape=25, dtype=np.uint32)

displacements = np.array( [12, 6, 3, 4], dtype=np.int32 )

comm.Gatherv(sendbuf=[x, mpi.DOUBLE], recvbuf=[X, (displacements, None)], root=0)
comm.Gatherv(sendbuf=[y, mpi.DOUBLE], recvbuf=[Y, (displacements, None)], root=0)
comm.Gatherv(sendbuf=[gid, mpi.UNSIGNED_INT], recvbuf=[GID, (displacements, None)], root=0)

# broadcast global X, Y and GID to everyone
comm.Bcast(buf=X, root=0)
comm.Bcast(buf=Y, root=0)
comm.Bcast(buf=GID, root=0)

# create the local particle arrays and exchange objects
pa = get_particle_array_wcsph(name='test', x=x, y=y, gid=gid)
pae = ParticleArrayExchange(pa_index=0, pa=pa, comm=comm)

# set the export indices for each array
pae.reset_lists()
pae.numParticleExport = numExport
pae.exportParticleLocalids.resize(numExport)
pae.exportParticleLocalids.set_data( exportLocalids )

pae.exportParticleProcs.resize( numExport )
pae.exportParticleProcs.set_data( exportProcs )

# call lb_balance with these lists
pae.lb_exchange_data()

# All arrays must be local after lb_exchange_data
assert( pa.num_real_particles == pa.get_number_of_particles() )
assert( np.allclose(pa.tag, 0) )

# now check the data on each array
numParticles = 6
if rank == 0:
    numParticles = 7

assert( pa.num_real_particles == numParticles )

for i in range(numParticles):
    assert( abs(X[pa.gid[i]] - pa.x[i]) < 1e-15 )
    assert( abs(Y[pa.gid[i]] - pa.y[i]) < 1e-15 )
    assert( GID[pa.gid[i]] == pa.gid[i] )