From ce028154a0e722b4e3e03d9a80dbf89b5a00bbd1 Mon Sep 17 00:00:00 2001
From: Abdullah Alperen <alperena@msu.edu>
Date: Sat, 15 Dec 2018 21:13:32 -0500
Subject: [PATCH] PuReMD: change timing functions with MPI_Wtime()
---
PuReMD/src/comm_tools.c | 4 +-
PuReMD/src/forces.c | 51 ++++++----
PuReMD/src/init_md.c | 4 +-
PuReMD/src/io_tools.c | 4 +-
PuReMD/src/linear_solvers.c | 197 ++++++++++++++++++++----------------
PuReMD/src/neighbors.c | 4 +-
PuReMD/src/parallelreax.c | 4 +-
PuReMD/src/qEq.c | 4 +-
PuReMD/src/reax_types.h | 2 +-
PuReMD/src/reset_tools.c | 2 +-
PuReMD/src/tool_box.c | 2 +
11 files changed, 156 insertions(+), 122 deletions(-)
diff --git a/PuReMD/src/comm_tools.c b/PuReMD/src/comm_tools.c
index 1227f63f..52b8c986 100644
--- a/PuReMD/src/comm_tools.c
+++ b/PuReMD/src/comm_tools.c
@@ -794,7 +794,7 @@ void Comm_Atoms( reax_system *system, control_params *control,
if ( system->my_rank == MASTER_NODE )
{
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
}
#endif
@@ -862,7 +862,7 @@ void Comm_Atoms( reax_system *system, control_params *control,
#if defined(LOG_PERFORMANCE)
if ( system->my_rank == MASTER_NODE )
{
- t_elapsed = Get_Timing_Info( t_start );
+ t_elapsed = MPI_Wtime() - t_start;
data->timing.comm += t_elapsed;
}
#endif
diff --git a/PuReMD/src/forces.c b/PuReMD/src/forces.c
index 21823d49..d6ba7625 100644
--- a/PuReMD/src/forces.c
+++ b/PuReMD/src/forces.c
@@ -379,7 +379,7 @@ void Init_Forces( reax_system *system, control_params *control,
renbr = (data->step - data->prev_steps) % control->reneighbor == 0;
#if defined(NEUTRAL_TERRITORY)
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
nt_flag = 1;
if( renbr )
{
@@ -421,7 +421,7 @@ void Init_Forces( reax_system *system, control_params *control,
mark[0] = mark[1] = 1;
mark[2] = mark[3] = mark[4] = mark[5] = 2;
- t_cm_init += Get_Timing_Info( t_start );
+ t_cm_init += MPI_Wtime() - t_start;
#endif
for ( i = 0; i < system->N; ++i )
@@ -466,12 +466,12 @@ void Init_Forces( reax_system *system, control_params *control,
ihb_top = -1;
if ( local == 1 )
{
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
H->start[i] = Htop;
H->entries[Htop].j = i;
H->entries[Htop].val = sbp_i->eta;
++Htop;
- t_cm_init += Get_Timing_Info( t_start );
+ t_cm_init += MPI_Wtime() - t_start;
if ( control->hbond_cut > 0 )
{
@@ -534,7 +534,7 @@ void Init_Forces( reax_system *system, control_params *control,
if ( local == 1 )
{
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
/* H matrix entry */
#if defined(NEUTRAL_TERRITORY)
if ( atom_j->nt_dir > 0 || (j < system->n
@@ -580,7 +580,7 @@ void Init_Forces( reax_system *system, control_params *control,
++Htop;
}
#endif
- t_cm_init += Get_Timing_Info( t_start );
+ t_cm_init += MPI_Wtime() - t_start;
/* hydrogen bond lists */
if ( control->hbond_cut > 0 && (ihb == 1 || ihb == 2) &&
@@ -612,7 +612,7 @@ void Init_Forces( reax_system *system, control_params *control,
#if defined(NEUTRAL_TERRITORY)
else if ( local == 2 )
{
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
/* H matrix entry */
if( ( atom_j->nt_dir != -1 && mark[atom_i->nt_dir] != mark[atom_j->nt_dir]
&& ( H->format == SYM_FULL_MATRIX
@@ -645,7 +645,7 @@ void Init_Forces( reax_system *system, control_params *control,
++Htop;
}
- t_cm_init += Get_Timing_Info( t_start );
+ t_cm_init += MPI_Wtime() - t_start;
}
#endif
@@ -670,7 +670,7 @@ void Init_Forces( reax_system *system, control_params *control,
}
Set_End_Index( i, btop_i, bonds );
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
if ( local == 1 )
{
H->end[i] = Htop;
@@ -691,7 +691,7 @@ void Init_Forces( reax_system *system, control_params *control,
}
}
#endif
- t_cm_init += Get_Timing_Info( t_start );
+ t_cm_init += MPI_Wtime() - t_start;
}
if ( far_nbrs->format == FULL_LIST )
@@ -725,11 +725,11 @@ void Init_Forces( reax_system *system, control_params *control,
}
}
- MPI_Reduce(&t_cm_init, &total_cm_init, 1, MPI_DOUBLE, MPI_SUM, MASTER_NODE, mpi_data->world);
+ /*MPI_Reduce(&t_cm_init, &total_cm_init, 1, MPI_DOUBLE, MPI_SUM, MASTER_NODE, mpi_data->world);
if( system->my_rank == MASTER_NODE )
{
data->timing.init_qeq += total_cm_init / control->nprocs;
- }
+ }*/
#if defined(DEBUG)
Print_Sparse_Matrix2( system, H, NULL );
@@ -1215,7 +1215,7 @@ void Compute_Forces( reax_system *system, control_params *control,
//MPI_Barrier( mpi_data->world );
if ( system->my_rank == MASTER_NODE )
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
#endif
comm = mpi_data->world;
@@ -1236,7 +1236,12 @@ void Compute_Forces( reax_system *system, control_params *control,
#if defined(LOG_PERFORMANCE)
//MPI_Barrier( mpi_data->world );
if ( system->my_rank == MASTER_NODE )
- Update_Timing_Info( &t_start, &(data->timing.init_forces) );
+ {
+ //Update_Timing_Info( &t_start, &(data->timing.init_forces) );
+ real t_end = MPI_Wtime();
+ data->timing.init_forces += t_end - t_start;
+ t_start = t_end;
+ }
#endif
@@ -1248,7 +1253,7 @@ void Compute_Forces( reax_system *system, control_params *control,
//MPI_Barrier( mpi_data->world );
if ( system->my_rank == MASTER_NODE )
{
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
}
#endif
#if defined(DEBUG_FOCUS)
@@ -1267,7 +1272,7 @@ void Compute_Forces( reax_system *system, control_params *control,
//MPI_Barrier( mpi_data->world );
if ( system->my_rank == MASTER_NODE )
{
- t_elapsed = Get_Timing_Info( t_start );
+ t_elapsed = MPI_Wtime() - t_start;
data->timing.cm += t_elapsed;
}
#endif
@@ -1285,7 +1290,12 @@ void Compute_Forces( reax_system *system, control_params *control,
#if defined(LOG_PERFORMANCE)
//MPI_Barrier( mpi_data->world );
if ( system->my_rank == MASTER_NODE )
- Update_Timing_Info( &t_start, &(data->timing.nonb) );
+ {
+ //Update_Timing_Info( &t_start, &(data->timing.nonb) );
+ real t_end = MPI_Wtime();
+ data->timing.nonb += t_end - t_start;
+ t_start = t_end;
+ }
#endif
#if defined(DEBUG_FOCUS)
fprintf( stderr, "p%d @ step%d: nonbonded forces completed\n",
@@ -1300,7 +1310,12 @@ void Compute_Forces( reax_system *system, control_params *control,
#if defined(LOG_PERFORMANCE)
//MPI_Barrier( mpi_data->world );
if ( system->my_rank == MASTER_NODE )
- Update_Timing_Info( &t_start, &(data->timing.bonded) );
+ {
+ //Update_Timing_Info( &t_start, &(data->timing.bonded) );
+ real t_end = MPI_Wtime();
+ data->timing.bonded += t_end - t_start;
+ t_start = t_end;
+ }
#endif
#if defined(DEBUG_FOCUS)
fprintf( stderr, "p%d @ step%d: total forces computed\n",
diff --git a/PuReMD/src/init_md.c b/PuReMD/src/init_md.c
index f3efe9d1..16e91bc8 100644
--- a/PuReMD/src/init_md.c
+++ b/PuReMD/src/init_md.c
@@ -280,7 +280,7 @@ int Init_Simulation_Data( reax_system *system, control_params *control,
MPI_Barrier( mpi_data->world ); // wait for everyone to come here
if ( system->my_rank == MASTER_NODE )
{
- data->timing.start = Get_Time( );
+ data->timing.start = MPI_Wtime();
#if defined(LOG_PERFORMANCE)
//Reset_Timing( &data->timing );
/* init timing info */
@@ -354,7 +354,7 @@ int Init_Simulation_Data( reax_system *system, control_params *control,
/* initialize the timer(s) */
if ( system->my_rank == MASTER_NODE )
{
- data->timing.start = Get_Time( );
+ data->timing.start = MPI_Wtime();
#if defined(LOG_PERFORMANCE)
//Reset_Timing( &data->timing );
/* init timing info */
diff --git a/PuReMD/src/io_tools.c b/PuReMD/src/io_tools.c
index cf62b27b..40e73405 100644
--- a/PuReMD/src/io_tools.c
+++ b/PuReMD/src/io_tools.c
@@ -1093,7 +1093,7 @@ void Output_Results( reax_system *system, control_params *control,
#endif //DEBUG
#if defined(LOG_PERFORMANCE)
- t_elapsed = Get_Timing_Info( data->timing.total );
+ t_elapsed = MPI_Wtime() - data->timing.total;
if ( data->step - data->prev_steps > 0 )
denom = 1.0 / out_control->energy_update_freq;
else denom = 1;
@@ -1119,7 +1119,7 @@ void Output_Results( reax_system *system, control_params *control,
data->timing.cm_solver_tri_solve * denom );
//Reset_Timing( &(data->timing) );
- data->timing.total = Get_Time( );
+ data->timing.total = MPI_Wtime();
data->timing.comm = ZERO;
data->timing.nbrs = 0;
data->timing.init_forces = 0;
diff --git a/PuReMD/src/linear_solvers.c b/PuReMD/src/linear_solvers.c
index 2ff188c2..4a7b258e 100644
--- a/PuReMD/src/linear_solvers.c
+++ b/PuReMD/src/linear_solvers.c
@@ -134,7 +134,7 @@ real setup_sparse_approx_inverse( reax_system *system, simulation_data *data, st
real start, t_start, t_comm;
real total_comm;
- start = Get_Time();
+ start = MPI_Wtime();
t_comm = 0.0;
srecv = NULL;
@@ -190,10 +190,10 @@ real setup_sparse_approx_inverse( reax_system *system, simulation_data *data, st
}
s_local = (int) (12.0 * log2(n_local * nprocs));
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Allreduce( &n_local, &n, 1, MPI_INT, MPI_SUM, comm );
MPI_Reduce( &s_local, &s, 1, MPI_INT, MPI_SUM, MASTER_NODE, comm );
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
/* count num. bin elements for each processor, uniform bin sizes */
input_array = smalloc( sizeof(real) * n_local,
@@ -240,9 +240,9 @@ real setup_sparse_approx_inverse( reax_system *system, simulation_data *data, st
}
/* gather samples at the root process */
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Gather( &s_local, 1, MPI_INT, srecv, 1, MPI_INT, MASTER_NODE, comm );
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
if( system->my_rank == MASTER_NODE )
{
@@ -253,10 +253,10 @@ real setup_sparse_approx_inverse( reax_system *system, simulation_data *data, st
}
}
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Gatherv( samplelist_local, s_local, MPI_DOUBLE,
samplelist, srecv, sdispls, MPI_DOUBLE, MASTER_NODE, comm);
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
/* sort samples at the root process and select pivots */
if ( system->my_rank == MASTER_NODE )
@@ -270,9 +270,9 @@ real setup_sparse_approx_inverse( reax_system *system, simulation_data *data, st
}
/* broadcast pivots */
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Bcast( pivotlist, nprocs - 1, MPI_DOUBLE, MASTER_NODE, comm );
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
for ( i = 0; i < nprocs; ++i )
{
@@ -308,9 +308,9 @@ real setup_sparse_approx_inverse( reax_system *system, simulation_data *data, st
}
/* determine counts for elements per process */
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Allreduce( MPI_IN_PLACE, scounts, nprocs, MPI_INT, MPI_SUM, comm );
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
/* find the target process */
target_proc = 0;
@@ -336,10 +336,10 @@ real setup_sparse_approx_inverse( reax_system *system, simulation_data *data, st
}
/* send local buckets to target processor for quickselect */
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Gather( scounts_local + target_proc, 1, MPI_INT, scounts,
1, MPI_INT, target_proc, comm );
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
if ( system->my_rank == target_proc )
{
@@ -350,10 +350,10 @@ real setup_sparse_approx_inverse( reax_system *system, simulation_data *data, st
}
}
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Gatherv( bucketlist_local + dspls_local[target_proc], scounts_local[target_proc], MPI_DOUBLE,
bucketlist, scounts, dspls, MPI_DOUBLE, target_proc, comm);
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
/* apply quick select algorithm at the target process */
if ( system->my_rank == target_proc )
@@ -421,9 +421,9 @@ real setup_sparse_approx_inverse( reax_system *system, simulation_data *data, st
}
/* broadcast the filtering value */
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Bcast( &threshold, 1, MPI_DOUBLE, target_proc, comm );
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
#if defined(DEBUG)
int nnz = 0; //uncomment to check the nnz's in the sparsity pattern
@@ -490,7 +490,7 @@ real setup_sparse_approx_inverse( reax_system *system, simulation_data *data, st
sfree( bucketlist, "setup_sparse_approx_inverse::bucketlist" );
}
- return Get_Timing_Info( start );
+ return MPI_Wtime() - start;
}
@@ -527,7 +527,7 @@ real sparse_approx_inverse( reax_system *system, simulation_data *data,
real start, t_start, t_comm;
real total_comm;
///////////////////
- start = Get_Time( );
+ start = MPI_Wtime();
t_comm = 0.0;
comm = mpi_data->world;
@@ -579,10 +579,10 @@ real sparse_approx_inverse( reax_system *system, simulation_data *data,
}
/* Announce the nnz's in each row that will be communicated later */
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
scale = sizeof(int) / sizeof(void);
Dist_NT( system, mpi_data, row_nnz, MPI_INT, scale, int_packer );
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
fprintf( stdout,"SAI after Dist_NT call\n");
fflush( stdout );
@@ -614,10 +614,10 @@ real sparse_approx_inverse( reax_system *system, simulation_data *data,
fprintf( stdout,"Dist_NT communication receive phase direction %d will receive %d\n", d, cnt);
fflush( stdout );
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Irecv( j_recv + d, cnt, MPI_INT, nbr->receive_rank, d, comm, &(req[2*d]) );
MPI_Irecv( val_recv + d, cnt, MPI_DOUBLE, nbr->receive_rank, d, comm, &(req[2*d+1]) );
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
}
}
}
@@ -662,14 +662,14 @@ real sparse_approx_inverse( reax_system *system, simulation_data *data,
fprintf( stdout,"Dist_NT communication send phase direction %d will send %d\n", d, cnt );
fflush( stdout );
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Send( j_send, cnt, MPI_INT, nbr->rank, d, comm );
fprintf( stdout,"Dist_NT communication send phase direction %d cnt = %d\n", d, cnt);
fflush( stdout );
MPI_Send( val_send, cnt, MPI_DOUBLE, nbr->rank, d, comm );
fprintf( stdout,"Dist_NT communication send phase direction %d cnt = %d\n", d, cnt);
fflush( stdout );
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
}
}
}
@@ -678,9 +678,9 @@ real sparse_approx_inverse( reax_system *system, simulation_data *data,
///////////////////////
for ( d = 0; d < count; ++d )
{
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Waitany( REAX_MAX_NT_NBRS, req, &index, stat);
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
nbr = &(system->my_nt_nbrs[index/2]);
cnt = 0;
@@ -881,7 +881,7 @@ real sparse_approx_inverse( reax_system *system, simulation_data *data,
data->timing.cm_solver_comm += total_comm / nprocs;
}
- return Get_Timing_Info( start );
+ return MPI_Wtime() - start;
}
#else
real sparse_approx_inverse( reax_system *system, simulation_data *data,
@@ -912,7 +912,7 @@ real sparse_approx_inverse( reax_system *system, simulation_data *data,
real start, t_start, t_comm;
real total_comm;
- start = Get_Time( );
+ start = MPI_Wtime();
t_comm = 0.0;
comm = mpi_data->world;
@@ -957,10 +957,10 @@ real sparse_approx_inverse( reax_system *system, simulation_data *data,
}
/* Announce the nnz's in each row that will be communicated later */
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
scale = sizeof(int) / sizeof(void);
Dist( system, mpi_data, row_nnz, MPI_INT, scale, int_packer );
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
comm = mpi_data->comm_mesh3D;
out_bufs = mpi_data->out_buffers;
@@ -996,10 +996,10 @@ real sparse_approx_inverse( reax_system *system, simulation_data *data,
val_recv1 = smalloc( sizeof(real) * cnt,
"sparse_approx_inverse::val_recv1", MPI_COMM_WORLD );
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Irecv( j_recv1, cnt, MPI_INT, nbr1->rank, 2 * d + 1, comm, &req1 );
MPI_Irecv( val_recv1, cnt, MPI_DOUBLE, nbr1->rank, 2 * d + 1, comm, &req2 );
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
}
}
@@ -1025,10 +1025,10 @@ real sparse_approx_inverse( reax_system *system, simulation_data *data,
val_recv2 = smalloc( sizeof(real) * cnt,
"sparse_approx_inverse::val_recv2", MPI_COMM_WORLD );
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Irecv( j_recv2, cnt, MPI_INT, nbr2->rank, 2 * d, comm, &req3 );
MPI_Irecv( val_recv2, cnt, MPI_DOUBLE, nbr2->rank, 2 * d, comm, &req4 );
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
}
}
@@ -1072,10 +1072,10 @@ real sparse_approx_inverse( reax_system *system, simulation_data *data,
}
}
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Send( j_send, cnt, MPI_INT, nbr1->rank, 2 * d, comm );
MPI_Send( val_send, cnt, MPI_DOUBLE, nbr1->rank, 2 * d, comm );
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
}
}
@@ -1118,20 +1118,20 @@ real sparse_approx_inverse( reax_system *system, simulation_data *data,
}
}
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Send( j_send, cnt, MPI_INT, nbr2->rank, 2 * d + 1, comm );
MPI_Send( val_send, cnt, MPI_DOUBLE, nbr2->rank, 2 * d + 1, comm );
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
}
}
if ( flag1 )
{
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Wait( &req1, &stat1 );
MPI_Wait( &req2, &stat2 );
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
cnt = 0;
for ( i = nbr1->atoms_str; i < (nbr1->atoms_str + nbr1->atoms_cnt); ++i )
@@ -1155,10 +1155,10 @@ real sparse_approx_inverse( reax_system *system, simulation_data *data,
if ( flag2 )
{
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
MPI_Wait( &req3, &stat3 );
MPI_Wait( &req4, &stat4 );
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
cnt = 0;
for ( i = nbr2->atoms_str; i < (nbr2->atoms_str + nbr2->atoms_cnt); ++i )
@@ -1368,7 +1368,7 @@ real sparse_approx_inverse( reax_system *system, simulation_data *data,
data->timing.cm_solver_comm += total_comm / nprocs;
}
- return Get_Timing_Info( start );
+ return MPI_Wtime() - start;
}
#endif
#endif
@@ -1450,7 +1450,7 @@ void dual_Sparse_MatVec( sparse_matrix *A, rvec2 *x, rvec2 *b, int N )
{
matvecs = 0;
t_start = matvec_time = dot_time = 0;
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
}
#endif
@@ -1464,7 +1464,10 @@ void dual_Sparse_MatVec( sparse_matrix *A, rvec2 *x, rvec2 *b, int N )
#if defined(CG_PERFORMANCE)
if ( system->my_rank == MASTER_NODE )
{
- Update_Timing_Info( &t_start, &matvec_time );
+ //Update_Timing_Info( &t_start, &matvec_time );
+ real t_end = MPI_Wtime();
+ matvec_time += t_end - t_start;
+ t_start = t_end;
}
#endif
@@ -1502,7 +1505,12 @@ void dual_Sparse_MatVec( sparse_matrix *A, rvec2 *x, rvec2 *b, int N )
#if defined(CG_PERFORMANCE)
if ( system->my_rank == MASTER_NODE )
- Update_Timing_Info( &t_start, &dot_time );
+ {
+ //Update_Timing_Info( &t_start, &dot_time );
+ real t_end = MPI_Wtime();
+ dot_time += t_end - t_start;
+ t_start = t_end;
+ }
#endif
for ( i = 1; i < 300; ++i )
@@ -1517,7 +1525,10 @@ void dual_Sparse_MatVec( sparse_matrix *A, rvec2 *x, rvec2 *b, int N )
#if defined(CG_PERFORMANCE)
if ( system->my_rank == MASTER_NODE )
{
- Update_Timing_Info( &t_start, &matvec_time );
+ //Update_Timing_Info( &t_start, &matvec_time );
+ real t_end = MPI_Wtime();
+ matvec_time += t_end - t_start;
+ t_start = t_end;
}
#endif
@@ -1557,7 +1568,10 @@ void dual_Sparse_MatVec( sparse_matrix *A, rvec2 *x, rvec2 *b, int N )
#if defined(CG_PERFORMANCE)
if ( system->my_rank == MASTER_NODE )
{
- Update_Timing_Info( &t_start, &dot_time );
+ //Update_Timing_Info( &t_start, &dot_time );
+ real t_end = MPI_Wtime();
+ dot_time += t_end - t_start;
+ t_start = t_end;
}
#endif
@@ -1737,164 +1751,168 @@ int CG( reax_system *system, control_params *control, simulation_data *data,
t_comm = 0.0;
t_allreduce = 0.0;
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
scale = sizeof(real) / sizeof(void);
#if defined(NEUTRAL_TERRITORY)
Dist_NT( system, mpi_data, x, MPI_DOUBLE, scale, real_packer );
#else
Dist( system, mpi_data, x, MPI_DOUBLE, scale, real_packer );
#endif
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
#if defined(NEUTRAL_TERRITORY)
Sparse_MatVec( H, x, workspace->q, H->NT );
#else
Sparse_MatVec( H, x, workspace->q, system->N );
#endif
- t_spmv += Get_Timing_Info( t_start );
+ t_spmv += MPI_Wtime() - t_start;
if ( H->format == SYM_HALF_MATRIX )
{
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
#if defined(NEUTRAL_TERRITORY)
Coll_NT( system, mpi_data, workspace->q, MPI_DOUBLE, scale, real_unpacker );
#else
Coll( system, mpi_data, workspace->q, MPI_DOUBLE, scale, real_unpacker );
#endif
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
}
else
{
+ t_start = MPI_Wtime();
#if defined(NEUTRAL_TERRITORY)
Coll_NT( system, mpi_data, workspace->q, MPI_DOUBLE, scale, real_unpacker );
#endif
+ t_comm += MPI_Wtime() - t_start;
}
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
Vector_Sum( workspace->r , 1., b, -1., workspace->q, system->n );
- t_vops += Get_Timing_Info( t_start );
+ t_vops += MPI_Wtime() - t_start;
/* pre-conditioning */
if( control->cm_solver_pre_comp_type == SAI_PC )
{
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
#if defined(NEUTRAL_TERRITORY)
Dist_NT( system, mpi_data, workspace->r, MPI_DOUBLE, scale, real_packer );
#else
Dist( system, mpi_data, workspace->r, MPI_DOUBLE, scale, real_packer );
#endif
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
#if defined(NEUTRAL_TERRITORY)
Sparse_MatVec( workspace->H_app_inv, workspace->r, workspace->d, H->NT );
#else
Sparse_MatVec( workspace->H_app_inv, workspace->r, workspace->d, system->n );
#endif
- t_pa += Get_Timing_Info( t_start );
+ t_pa += MPI_Wtime() - t_start;
}
else if ( control->cm_solver_pre_comp_type == JACOBI_PC)
{
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
for ( j = 0; j < system->n; ++j )
{
workspace->d[j] = workspace->r[j] * workspace->Hdia_inv[j];
}
- t_pa += Get_Timing_Info( t_start );
+ t_pa += MPI_Wtime() - t_start;
}
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
b_norm = Parallel_Norm( b, system->n, mpi_data->world );
sig_new = Parallel_Dot(workspace->r, workspace->d, system->n, mpi_data->world);
- t_allreduce += Get_Timing_Info( t_start );
+ t_allreduce += MPI_Wtime() - t_start;
for ( i = 0; i < control->cm_solver_max_iters && sqrt(sig_new) / b_norm > tol; ++i )
{
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
#if defined(NEUTRAL_TERRITORY)
Dist_NT( system, mpi_data, workspace->d, MPI_DOUBLE, scale, real_packer );
#else
Dist( system, mpi_data, workspace->d, MPI_DOUBLE, scale, real_packer );
#endif
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
#if defined(NEUTRAL_TERRITORY)
Sparse_MatVec( H, workspace->d, workspace->q, H->NT );
#else
Sparse_MatVec( H, workspace->d, workspace->q, system->N );
#endif
- t_spmv += Get_Timing_Info( t_start );
+ t_spmv += MPI_Wtime() - t_start;
if ( H->format == SYM_HALF_MATRIX )
{
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
#if defined(NEUTRAL_TERRITORY)
Coll_NT(system, mpi_data, workspace->q, MPI_DOUBLE, scale, real_unpacker);
#else
Coll(system, mpi_data, workspace->q, MPI_DOUBLE, scale, real_unpacker);
#endif
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
}
else
{
+ t_start = MPI_Wtime();
#if defined(NEUTRAL_TERRITORY)
Coll_NT( system, mpi_data, workspace->q, MPI_DOUBLE, scale, real_unpacker );
#endif
+ t_comm += MPI_Wtime() - t_start;
}
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
tmp = Parallel_Dot(workspace->d, workspace->q, system->n, mpi_data->world);
- t_allreduce += Get_Timing_Info ( t_start );
+ t_allreduce += MPI_Wtime() - t_start;
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
alpha = sig_new / tmp;
Vector_Add( x, alpha, workspace->d, system->n );
Vector_Add( workspace->r, -alpha, workspace->q, system->n );
- t_vops += Get_Timing_Info( t_start );
+ t_vops += MPI_Wtime() - t_start;
/* pre-conditioning */
if( control->cm_solver_pre_comp_type == SAI_PC )
{
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
#if defined(NEUTRAL_TERRITORY)
Dist_NT( system, mpi_data, workspace->r, MPI_DOUBLE, scale, real_packer );
#else
Dist( system, mpi_data, workspace->r, MPI_DOUBLE, scale, real_packer );
#endif
- t_comm += Get_Timing_Info( t_start );
+ t_comm += MPI_Wtime() - t_start;
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
#if defined(NEUTRAL_TERRITORY)
Sparse_MatVec( workspace->H_app_inv, workspace->r, workspace->p, H->NT );
#else
Sparse_MatVec( workspace->H_app_inv, workspace->r, workspace->p, system->n );
#endif
- t_pa += Get_Timing_Info( t_start );
+ t_pa += MPI_Wtime() - t_start;
}
else if ( control->cm_solver_pre_comp_type == JACOBI_PC)
{
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
for ( j = 0; j < system->n; ++j )
{
workspace->p[j] = workspace->r[j] * workspace->Hdia_inv[j];
}
- t_pa += Get_Timing_Info( t_start );
+ t_pa += MPI_Wtime() - t_start;
}
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
sig_old = sig_new;
sig_new = Parallel_Dot(workspace->r, workspace->p, system->n, mpi_data->world);
- t_allreduce += Get_Timing_Info( t_start );
+ t_allreduce += MPI_Wtime() - t_start;
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
beta = sig_new / sig_old;
Vector_Sum( workspace->d, 1., workspace->p, beta, workspace->d, system->n );
- t_vops += Get_Timing_Info( t_start );
+ t_vops += MPI_Wtime() - t_start;
}
MPI_Reduce(&t_pa, &total_pa, 1, MPI_DOUBLE, MPI_SUM, MASTER_NODE, mpi_data->world);
@@ -1914,7 +1932,6 @@ int CG( reax_system *system, control_params *control, simulation_data *data,
MPI_Barrier(mpi_data->world);
- //TODO only master node should report it
if ( i >= control->cm_solver_max_iters && system->my_rank == MASTER_NODE )
{
fprintf( stderr, "CG convergence failed!\n" );
@@ -1970,7 +1987,7 @@ int CG( reax_system *system, control_params *control, simulation_data *data,
{
#if defined(CG_PERFORMANCE)
if ( system->my_rank == MASTER_NODE )
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
#endif
Dist( system, mpi_data, workspace->d, MPI_DOUBLE, scale, real_packer );
Sparse_MatVec( H, workspace->d, workspace->q, system->N );
@@ -1979,14 +1996,14 @@ int CG( reax_system *system, control_params *control, simulation_data *data,
#if defined(CG_PERFORMANCE)
if ( system->my_rank == MASTER_NODE )
{
- t_elapsed = Get_Timing_Info( t_start );
+ t_elapsed = MPI_Wtime() - t_start;
matvec_time += t_elapsed;
}
#endif
#if defined(CG_PERFORMANCE)
if ( system->my_rank == MASTER_NODE )
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
#endif
tmp = Parallel_Dot(workspace->d, workspace->q, system->n, mpi_data->world);
alpha = sig_new / tmp;
@@ -2026,7 +2043,7 @@ int CG( reax_system *system, control_params *control, simulation_data *data,
#if defined(CG_PERFORMANCE)
if ( system->my_rank == MASTER_NODE )
{
- t_elapsed = Get_Timing_Info( t_start );
+ t_elapsed = MPI_Wtime() - t_start;
dot_time += t_elapsed;
}
#endif
diff --git a/PuReMD/src/neighbors.c b/PuReMD/src/neighbors.c
index 0bd33ea8..e59269ab 100644
--- a/PuReMD/src/neighbors.c
+++ b/PuReMD/src/neighbors.c
@@ -81,7 +81,7 @@ void Generate_Neighbor_Lists( reax_system *system, simulation_data *data,
real t_start = 0, t_elapsed = 0;
if ( system->my_rank == MASTER_NODE )
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
#endif
// fprintf( stderr, "\n\tentered nbrs - " );
@@ -166,7 +166,7 @@ void Generate_Neighbor_Lists( reax_system *system, simulation_data *data,
#if defined(LOG_PERFORMANCE)
if ( system->my_rank == MASTER_NODE )
{
- t_elapsed = Get_Timing_Info( t_start );
+ t_elapsed = MPI_Wtime() - t_start;
data->timing.nbrs += t_elapsed;
}
#endif
diff --git a/PuReMD/src/parallelreax.c b/PuReMD/src/parallelreax.c
index 176e1484..0ff49df1 100644
--- a/PuReMD/src/parallelreax.c
+++ b/PuReMD/src/parallelreax.c
@@ -186,7 +186,7 @@ int main( int argc, char* argv[] )
/* measure total simulation time after input is read */
if ( system->my_rank == MASTER_NODE )
- t_start = Get_Time( );
+ t_start = MPI_Wtime();
/* initialize datastructures */
Initialize( system, control, data, workspace, lists, out_control, mpi_data );
@@ -248,7 +248,7 @@ int main( int argc, char* argv[] )
/* end of the simulation, write total simulation time */
if ( system->my_rank == MASTER_NODE )
{
- t_elapsed = Get_Timing_Info( t_start );
+ t_elapsed = MPI_Wtime() - t_start;
fprintf( out_control->out, "Total Simulation Time: %.2f secs\n", t_elapsed );
fprintf( out_control->log, "Avg. # of Solver Itrs: %.2f\n", total_itr/((double)control->nsteps) );
}
diff --git a/PuReMD/src/qEq.c b/PuReMD/src/qEq.c
index 29cccf43..a29896f8 100644
--- a/PuReMD/src/qEq.c
+++ b/PuReMD/src/qEq.c
@@ -376,9 +376,9 @@ static void Setup_Preconditioner_QEq( reax_system *system, control_params *contr
real time, t_sort, t_pc, total_sort, total_pc;
/* sort H needed for SpMV's in linear solver, H or H_sp needed for preconditioning */
- time = Get_Time( );
+ time = MPI_Wtime();
Sort_Matrix_Rows( workspace->H );
- t_sort = Get_Timing_Info( time );
+ t_sort = MPI_Wtime() - time;
t_pc = setup_sparse_approx_inverse( system, data, workspace, mpi_data, workspace->H, &workspace->H_spar_patt,
control->nprocs, control->cm_solver_pre_comp_sai_thres );
diff --git a/PuReMD/src/reax_types.h b/PuReMD/src/reax_types.h
index 644fceae..c8ed1a71 100644
--- a/PuReMD/src/reax_types.h
+++ b/PuReMD/src/reax_types.h
@@ -40,7 +40,7 @@
/************* SOME DEFS - crucial for reax_types.h *********/
#define PURE_REAX
-#define NEUTRAL_TERRITORY
+//#define NEUTRAL_TERRITORY
//#define LAMMPS_REAX
//#define DEBUG
//#define DEBUG_FOCUS
diff --git a/PuReMD/src/reset_tools.c b/PuReMD/src/reset_tools.c
index 32811672..d4d1e6e7 100644
--- a/PuReMD/src/reset_tools.c
+++ b/PuReMD/src/reset_tools.c
@@ -106,7 +106,7 @@ void Reset_Simulation_Data( simulation_data* data, int virial )
void Reset_Timing( reax_timing *rt )
{
- /*rt->total = Get_Time();
+ /*rt->total = MPI_Wtime();
rt->comm = 0;
rt->nbrs = 0;
rt->init_forces = 0;
diff --git a/PuReMD/src/tool_box.c b/PuReMD/src/tool_box.c
index d898f219..9374fc3c 100644
--- a/PuReMD/src/tool_box.c
+++ b/PuReMD/src/tool_box.c
@@ -318,6 +318,8 @@ void Trim_Spaces( char *element )
struct timeval tim;
real t_end;
+// NOTE: these timing functions are not being used
+// replaced by MPI_Wtime()
real Get_Time( )
{
gettimeofday(&tim, NULL );
--
GitLab