diff --git a/examples/COUPLE/plugin/README b/examples/COUPLE/plugin/README index b035ed3c0e..bced08b6d7 100644 --- a/examples/COUPLE/plugin/README +++ b/examples/COUPLE/plugin/README @@ -13,7 +13,7 @@ like below. mpicc -c -O -Wall -g -I$HOME/lammps/src liblammpsplugin.c mpicc -c -O -Wall -g simple.c -mpicc simple.o liblammsplugin.o -ldl -o simpleC +mpicc simple.o liblammpsplugin.o -ldl -o simpleC You also need to build LAMMPS as a shared library (see examples/COUPLE/README), e.g. diff --git a/examples/COUPLE/plugin/liblammpsplugin.c b/examples/COUPLE/plugin/liblammpsplugin.c index 474b0a86c2..ea2e2ab536 100644 --- a/examples/COUPLE/plugin/liblammpsplugin.c +++ b/examples/COUPLE/plugin/liblammpsplugin.c @@ -31,51 +31,105 @@ liblammpsplugin_t *liblammpsplugin_load(const char *lib) if (lib == NULL) return NULL; handle = dlopen(lib,RTLD_NOW|RTLD_GLOBAL); if (handle == NULL) return NULL; - + lmp = (liblammpsplugin_t *) malloc(sizeof(liblammpsplugin_t)); lmp->handle = handle; #define ADDSYM(symbol) lmp->symbol = dlsym(handle,"lammps_" #symbol) ADDSYM(open); ADDSYM(open_no_mpi); + ADDSYM(open_fortran); ADDSYM(close); - ADDSYM(version); + + ADDSYM(mpi_init); + ADDSYM(mpi_finalize); + ADDSYM(kokkos_finalize); + ADDSYM(python_finalize); + ADDSYM(file); ADDSYM(command); ADDSYM(commands_list); ADDSYM(commands_string); - ADDSYM(free); - ADDSYM(extract_setting); - ADDSYM(extract_global); + + ADDSYM(get_natoms); + ADDSYM(get_thermo); + ADDSYM(extract_box); + ADDSYM(reset_box); + + ADDSYM(memory_usage); + ADDSYM(get_mpi_comm); + + ADDSYM(extract_setting); + ADDSYM(extract_global_datatype); + ADDSYM(extract_global); + + ADDSYM(extract_atom_datatype); ADDSYM(extract_atom); + ADDSYM(extract_compute); ADDSYM(extract_fix); ADDSYM(extract_variable); - - ADDSYM(get_thermo); - ADDSYM(get_natoms); - ADDSYM(set_variable); - ADDSYM(reset_box); ADDSYM(gather_atoms); ADDSYM(gather_atoms_concat); ADDSYM(gather_atoms_subset); ADDSYM(scatter_atoms); ADDSYM(scatter_atoms_subset); + ADDSYM(gather_bonds); - ADDSYM(set_fix_external_callback); + ADDSYM(create_atoms); - ADDSYM(config_has_package); - ADDSYM(config_package_count); - ADDSYM(config_package_name); + ADDSYM(find_pair_neighlist); + ADDSYM(find_fix_neighlist); + ADDSYM(find_compute_neighlist); + ADDSYM(neighlist_num_elements); + ADDSYM(neighlist_element_neighbors); + + ADDSYM(version); + ADDSYM(get_os_info); + + ADDSYM(config_has_mpi_support); ADDSYM(config_has_gzip_support); ADDSYM(config_has_png_support); ADDSYM(config_has_jpeg_support); ADDSYM(config_has_ffmpeg_support); ADDSYM(config_has_exceptions); - ADDSYM(create_atoms); + + ADDSYM(config_has_package); + ADDSYM(config_package_count); + ADDSYM(config_package_name); + + ADDSYM(config_accelerator); + ADDSYM(has_gpu_device); + ADDSYM(get_gpu_device_info); + + ADDSYM(has_style); + ADDSYM(style_count); + ADDSYM(style_name); + + ADDSYM(has_id); + ADDSYM(id_count); + ADDSYM(id_name); + + ADDSYM(plugin_count); + ADDSYM(plugin_name); + + ADDSYM(set_fix_external_callback); + ADDSYM(fix_external_get_force); + ADDSYM(fix_external_set_energy_global); + ADDSYM(fix_external_set_energy_peratom); + ADDSYM(fix_external_set_virial_global); + ADDSYM(fix_external_set_virial_peratom); + ADDSYM(fix_external_set_vector_length); + ADDSYM(fix_external_set_vector); + + ADDSYM(free); + + ADDSYM(is_running); + ADDSYM(force_timeout); + #ifdef LAMMPS_EXCEPTIONS lmp->has_exceptions = 1; ADDSYM(has_error); diff --git a/examples/COUPLE/plugin/liblammpsplugin.h b/examples/COUPLE/plugin/liblammpsplugin.h index c70316d8e6..219ddb2574 100644 --- a/examples/COUPLE/plugin/liblammpsplugin.h +++ b/examples/COUPLE/plugin/liblammpsplugin.h @@ -39,75 +39,121 @@ extern "C" { #if defined(LAMMPS_BIGBIG) typedef void (*FixExternalFnPtr)(void *, int64_t, int, int64_t *, double **, double **); -#elif defined(LAMMPS_SMALLBIG) -typedef void (*FixExternalFnPtr)(void *, int64_t, int, int *, double **, double **); -#else +#elif defined(LAMMPS_SMALLSMALL) typedef void (*FixExternalFnPtr)(void *, int, int, int *, double **, double **); +#else +typedef void (*FixExternalFnPtr)(void *, int64_t, int, int *, double **, double **); #endif - struct _liblammpsplugin { int abiversion; int has_exceptions; void *handle; - void (*open)(int, char **, MPI_Comm, void **); - void (*open_no_mpi)(int, char **, void **); + void *(*open)(int, char **, MPI_Comm, void **); + void *(*open_no_mpi)(int, char **, void **); + void *(*open_fortran)(int, char **, void **, int); void (*close)(void *); - int (*version)(void *); + + void (*mpi_init)(); + void (*mpi_finalize)(); + void (*kokkos_finalize)(); + void (*python_finalize)(); + void (*file)(void *, char *); - char *(*command)(void *, char *); - void (*commands_list)(void *, int, char **); - void (*commands_string)(void *, char *); - void (*free)(void *); - int (*extract_setting)(void *, char *); - void *(*extract_global)(void *, char *); - void (*extract_box)(void *, double *, double *, - double *, double *, double *, int *, int *); - void *(*extract_atom)(void *, char *); - void *(*extract_compute)(void *, char *, int, int); - void *(*extract_fix)(void *, char *, int, int, int, int); - void *(*extract_variable)(void *, char *, char *); + char *(*command)(void *, const char *); + void (*commands_list)(void *, int, const char **); + void (*commands_string)(void *, const char *); + double (*get_natoms)(void *); double (*get_thermo)(void *, char *); - int (*get_natoms)(void *); - int (*set_variable)(void *, char *, char *); + void (*extract_box)(void *, double *, double *, + double *, double *, double *, int *, int *); void (*reset_box)(void *, double *, double *, double, double, double); + void (*memory_usage)(void *, double *); + int (*get_mpi_comm)(void *); + + int (*extract_setting)(void *, const char *); + int *(*extract_global_datatype)(void *, const char *); + void *(*extract_global)(void *, const char *); + + void *(*extract_atom_datatype)(void *, const char *); + void *(*extract_atom)(void *, const char *); + + void *(*extract_compute)(void *, const char *, int, int); + void *(*extract_fix)(void *, const char *, int, int, int, int); + void *(*extract_variable)(void *, const char *, char *); + int (*set_variable)(void *, char *, char *); + void (*gather_atoms)(void *, char *, int, int, void *); void (*gather_atoms_concat)(void *, char *, int, int, void *); void (*gather_atoms_subset)(void *, char *, int, int, int, int *, void *); void (*scatter_atoms)(void *, char *, int, int, void *); void (*scatter_atoms_subset)(void *, char *, int, int, int, int *, void *); - void (*set_fix_external_callback)(void *, char *, FixExternalFnPtr, void*); + void (*gather_bonds)(void *, void *); + +// lammps_create_atoms() takes tagint and imageint as args +// ifdef insures they are compatible with rest of LAMMPS +// caller must match to how LAMMPS library is built - int (*config_has_package)(char * package_name); - int (*config_package_count)(); - int (*config_package_name)(int index, char * buffer, int max_size); +#ifndef LAMMPS_BIGBIG + void (*create_atoms)(void *, int, int *, int *, double *, + double *, int *, int); +#else + void (*create_atoms)(void *, int, int64_t *, int *, double *, + double *, int64_t *, int); +#endif + + int (*find_pair_neighlist)(void *, const char *, int, int, int); + int (*find_fix_neighlist)(void *, const char *, int); + int (*find_compute_neighlist)(void *, char *, int); + int (*neighlist_num_elements)(void *, int); + void (*neighlist_element_neighbors)(void *, int, int, int *, int *, int **); + + int (*version)(void *); + void (*get_os_info)(char *, int); + + int (*config_has_mpi_support)(); int (*config_has_gzip_support)(); int (*config_has_png_support)(); int (*config_has_jpeg_support)(); int (*config_has_ffmpeg_support)(); int (*config_has_exceptions)(); - int (*find_pair_neighlist)(void* ptr, char * style, int exact, int nsub, int request); - int (*find_fix_neighlist)(void* ptr, char * id, int request); - int (*find_compute_neighlist)(void* ptr, char * id, int request); - int (*neighlist_num_elements)(void* ptr, int idx); - void (*neighlist_element_neighbors)(void * ptr, int idx, int element, int * iatom, int * numneigh, int ** neighbors); + int (*config_has_package)(const char *); + int (*config_package_count)(); + int (*config_package_name)(int, char *, int); -// lammps_create_atoms() takes tagint and imageint as args -// ifdef insures they are compatible with rest of LAMMPS -// caller must match to how LAMMPS library is built + int (*config_accelerator)(const char *, const char *, const char *); + int (*has_gpu_device)(); + void (*get_gpu_device_info)(char *, int); -#ifdef LAMMPS_BIGBIG - void (*create_atoms)(void *, int, int64_t *, int *, - double *, double *, int64_t *, int); -#else - void (*create_atoms)(void *, int, int *, int *, - double *, double *, int *, int); -#endif + int (*has_style)(void *, const char *, const char *); + int (*style_count)(void *, const char *); + int (*style_name)(void *, const char *, int, char *, int); + + int (*has_id)(void *, const char *, const char *); + int (*id_count)(void *, const char *); + int (*id_name)(void *, const char *, int, char *, int); + + int (*plugin_count)(); + int (*plugin_name)(int, char *, char *, int); + + void (*set_fix_external_callback)(void *, const char *, FixExternalFnPtr, void*); + void (*fix_external_get_force)(void *, const char *); + void (*fix_external_set_energy_global)(void *, const char *, double); + void (*fix_external_set_energy_peratom)(void *, const char *, double *); + void (*fix_external_set_virial_global)(void *, const char *, double *); + void (*fix_external_set_virial_peratom)(void *, const char *, double **); + void (*fix_external_set_vector_length)(void *, const char *, int); + void (*fix_external_set_vector)(void *, const char *, int, double); + + void (*free)(void *); + + void (*is_running)(void *); + void (*force_timeout)(void *); int (*has_error)(void *); int (*get_last_error_message)(void *, char *, int); @@ -117,7 +163,7 @@ typedef struct _liblammpsplugin liblammpsplugin_t; liblammpsplugin_t *liblammpsplugin_load(const char *); int liblammpsplugin_release(liblammpsplugin_t *); - + #undef LAMMPS #ifdef __cplusplus } diff --git a/examples/COUPLE/plugin/log.simple.plugin.1 b/examples/COUPLE/plugin/log.simple.plugin.1 index ba4be378f4..0443e3ec17 100644 --- a/examples/COUPLE/plugin/log.simple.plugin.1 +++ b/examples/COUPLE/plugin/log.simple.plugin.1 @@ -1,9 +1,12 @@ -LAMMPS (18 Feb 2020) -Lattice spacing in x,y,z = 1.6796 1.6796 1.6796 -Created orthogonal box = (0 0 0) to (6.71838 6.71838 6.71838) +LAMMPS (31 Aug 2021) +OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (src/comm.cpp:98) + using 1 OpenMP thread(s) per MPI task +Lattice spacing in x,y,z = 1.6795962 1.6795962 1.6795962 +Created orthogonal box = (0.0000000 0.0000000 0.0000000) to (6.7183848 6.7183848 6.7183848) 1 by 1 by 1 MPI processor grid Created 256 atoms - create_atoms CPU = 0.000297844 secs + using lattice units in orthogonal box = (0.0000000 0.0000000 0.0000000) to (6.7183848 6.7183848 6.7183848) + create_atoms CPU = 0.001 seconds Neighbor list info ... update every 20 steps, delay 0 steps, check no max neighbors/atom: 2000, page size: 100000 @@ -14,108 +17,108 @@ Neighbor list info ... (1) pair lj/cut, perpetual attributes: half, newton on pair build: half/bin/atomonly/newton - stencil: half/bin/3d/newton + stencil: half/bin/3d bin: standard Setting up Verlet run ... Unit style : lj Current step : 0 Time step : 0.005 -Per MPI rank memory allocation (min/avg/max) = 2.63 | 2.63 | 2.63 Mbytes +Per MPI rank memory allocation (min/avg/max) = 2.630 | 2.630 | 2.630 Mbytes Step Temp E_pair E_mol TotEng Press 0 1.44 -6.7733681 0 -4.6218056 -5.0244179 10 1.1298532 -6.3095502 0 -4.6213906 -2.6058175 -Loop time of 0.00164276 on 1 procs for 10 steps with 256 atoms +Loop time of 0.00239712 on 1 procs for 10 steps with 256 atoms -Performance: 2629719.113 tau/day, 6087.313 timesteps/s -93.7% CPU use with 1 MPI tasks x no OpenMP threads +Performance: 1802163.347 tau/day, 4171.674 timesteps/s +97.2% CPU use with 1 MPI tasks x 1 OpenMP threads MPI task timing breakdown: Section | min time | avg time | max time |%varavg| %total --------------------------------------------------------------- -Pair | 0.0014956 | 0.0014956 | 0.0014956 | 0.0 | 91.04 +Pair | 0.0020572 | 0.0020572 | 0.0020572 | 0.0 | 85.82 Neigh | 0 | 0 | 0 | 0.0 | 0.00 -Comm | 8.045e-05 | 8.045e-05 | 8.045e-05 | 0.0 | 4.90 -Output | 1.1399e-05 | 1.1399e-05 | 1.1399e-05 | 0.0 | 0.69 -Modify | 3.7431e-05 | 3.7431e-05 | 3.7431e-05 | 0.0 | 2.28 -Other | | 1.789e-05 | | | 1.09 +Comm | 0.00018731 | 0.00018731 | 0.00018731 | 0.0 | 7.81 +Output | 4.478e-05 | 4.478e-05 | 4.478e-05 | 0.0 | 1.87 +Modify | 6.3637e-05 | 6.3637e-05 | 6.3637e-05 | 0.0 | 2.65 +Other | | 4.419e-05 | | | 1.84 -Nlocal: 256 ave 256 max 256 min +Nlocal: 256.000 ave 256 max 256 min Histogram: 1 0 0 0 0 0 0 0 0 0 -Nghost: 1431 ave 1431 max 1431 min +Nghost: 1431.00 ave 1431 max 1431 min Histogram: 1 0 0 0 0 0 0 0 0 0 -Neighs: 9984 ave 9984 max 9984 min +Neighs: 9984.00 ave 9984 max 9984 min Histogram: 1 0 0 0 0 0 0 0 0 0 Total # of neighbors = 9984 -Ave neighs/atom = 39 +Ave neighs/atom = 39.000000 Neighbor list builds = 0 Dangerous builds not checked Setting up Verlet run ... Unit style : lj Current step : 10 Time step : 0.005 -Per MPI rank memory allocation (min/avg/max) = 2.63 | 2.63 | 2.63 Mbytes +Per MPI rank memory allocation (min/avg/max) = 2.630 | 2.630 | 2.630 Mbytes Step Temp E_pair E_mol TotEng Press 10 1.1298532 -6.3095502 0 -4.6213906 -2.6058175 20 0.6239063 -5.557644 0 -4.6254403 0.97451173 -Loop time of 0.00199768 on 1 procs for 10 steps with 256 atoms +Loop time of 0.00329271 on 1 procs for 10 steps with 256 atoms -Performance: 2162504.180 tau/day, 5005.797 timesteps/s -99.8% CPU use with 1 MPI tasks x no OpenMP threads +Performance: 1311987.619 tau/day, 3037.008 timesteps/s +96.4% CPU use with 1 MPI tasks x 1 OpenMP threads MPI task timing breakdown: Section | min time | avg time | max time |%varavg| %total --------------------------------------------------------------- -Pair | 0.0018518 | 0.0018518 | 0.0018518 | 0.0 | 92.70 +Pair | 0.0029015 | 0.0029015 | 0.0029015 | 0.0 | 88.12 Neigh | 0 | 0 | 0 | 0.0 | 0.00 -Comm | 7.9768e-05 | 7.9768e-05 | 7.9768e-05 | 0.0 | 3.99 -Output | 1.1433e-05 | 1.1433e-05 | 1.1433e-05 | 0.0 | 0.57 -Modify | 3.6904e-05 | 3.6904e-05 | 3.6904e-05 | 0.0 | 1.85 -Other | | 1.773e-05 | | | 0.89 +Comm | 0.00021807 | 0.00021807 | 0.00021807 | 0.0 | 6.62 +Output | 4.9163e-05 | 4.9163e-05 | 4.9163e-05 | 0.0 | 1.49 +Modify | 7.0573e-05 | 7.0573e-05 | 7.0573e-05 | 0.0 | 2.14 +Other | | 5.339e-05 | | | 1.62 -Nlocal: 256 ave 256 max 256 min +Nlocal: 256.000 ave 256 max 256 min Histogram: 1 0 0 0 0 0 0 0 0 0 -Nghost: 1431 ave 1431 max 1431 min +Nghost: 1431.00 ave 1431 max 1431 min Histogram: 1 0 0 0 0 0 0 0 0 0 -Neighs: 9952 ave 9952 max 9952 min +Neighs: 9952.00 ave 9952 max 9952 min Histogram: 1 0 0 0 0 0 0 0 0 0 Total # of neighbors = 9952 -Ave neighs/atom = 38.875 +Ave neighs/atom = 38.875000 Neighbor list builds = 0 Dangerous builds not checked Setting up Verlet run ... Unit style : lj Current step : 20 Time step : 0.005 -Per MPI rank memory allocation (min/avg/max) = 2.63 | 2.63 | 2.63 Mbytes +Per MPI rank memory allocation (min/avg/max) = 2.630 | 2.630 | 2.630 Mbytes Step Temp E_pair E_mol TotEng Press 20 0.6239063 -5.5404291 0 -4.6082254 1.0394285 21 0.63845863 -5.5628733 0 -4.6089263 0.99398278 -Loop time of 0.000304321 on 1 procs for 1 steps with 256 atoms +Loop time of 0.000638039 on 1 procs for 1 steps with 256 atoms -Performance: 1419553.695 tau/day, 3286.004 timesteps/s -98.9% CPU use with 1 MPI tasks x no OpenMP threads +Performance: 677074.599 tau/day, 1567.302 timesteps/s +98.9% CPU use with 1 MPI tasks x 1 OpenMP threads MPI task timing breakdown: Section | min time | avg time | max time |%varavg| %total --------------------------------------------------------------- -Pair | 0.00027815 | 0.00027815 | 0.00027815 | 0.0 | 91.40 +Pair | 0.00042876 | 0.00042876 | 0.00042876 | 0.0 | 67.20 Neigh | 0 | 0 | 0 | 0.0 | 0.00 -Comm | 8.321e-06 | 8.321e-06 | 8.321e-06 | 0.0 | 2.73 -Output | 1.0513e-05 | 1.0513e-05 | 1.0513e-05 | 0.0 | 3.45 -Modify | 3.968e-06 | 3.968e-06 | 3.968e-06 | 0.0 | 1.30 -Other | | 3.365e-06 | | | 1.11 +Comm | 5.2872e-05 | 5.2872e-05 | 5.2872e-05 | 0.0 | 8.29 +Output | 0.00012218 | 0.00012218 | 0.00012218 | 0.0 | 19.15 +Modify | 1.3762e-05 | 1.3762e-05 | 1.3762e-05 | 0.0 | 2.16 +Other | | 2.047e-05 | | | 3.21 -Nlocal: 256 ave 256 max 256 min +Nlocal: 256.000 ave 256 max 256 min Histogram: 1 0 0 0 0 0 0 0 0 0 -Nghost: 1431 ave 1431 max 1431 min +Nghost: 1431.00 ave 1431 max 1431 min Histogram: 1 0 0 0 0 0 0 0 0 0 -Neighs: 9705 ave 9705 max 9705 min +Neighs: 9705.00 ave 9705 max 9705 min Histogram: 1 0 0 0 0 0 0 0 0 0 Total # of neighbors = 9705 -Ave neighs/atom = 37.9102 +Ave neighs/atom = 37.910156 Neighbor list builds = 0 Dangerous builds not checked Force on 1 atom via extract_atom: 26.9581 @@ -124,136 +127,136 @@ Setting up Verlet run ... Unit style : lj Current step : 21 Time step : 0.005 -Per MPI rank memory allocation (min/avg/max) = 2.63 | 2.63 | 2.63 Mbytes +Per MPI rank memory allocation (min/avg/max) = 2.630 | 2.630 | 2.630 Mbytes Step Temp E_pair E_mol TotEng Press 21 0.63845863 -5.5628733 0 -4.6089263 0.99398278 31 0.7494946 -5.7306417 0 -4.6107913 0.41043597 -Loop time of 0.00196027 on 1 procs for 10 steps with 256 atoms +Loop time of 0.00281277 on 1 procs for 10 steps with 256 atoms -Performance: 2203779.175 tau/day, 5101.341 timesteps/s -99.7% CPU use with 1 MPI tasks x no OpenMP threads +Performance: 1535852.558 tau/day, 3555.214 timesteps/s +92.6% CPU use with 1 MPI tasks x 1 OpenMP threads MPI task timing breakdown: Section | min time | avg time | max time |%varavg| %total --------------------------------------------------------------- -Pair | 0.0018146 | 0.0018146 | 0.0018146 | 0.0 | 92.57 +Pair | 0.0024599 | 0.0024599 | 0.0024599 | 0.0 | 87.45 Neigh | 0 | 0 | 0 | 0.0 | 0.00 -Comm | 8.0268e-05 | 8.0268e-05 | 8.0268e-05 | 0.0 | 4.09 -Output | 1.0973e-05 | 1.0973e-05 | 1.0973e-05 | 0.0 | 0.56 -Modify | 3.6913e-05 | 3.6913e-05 | 3.6913e-05 | 0.0 | 1.88 -Other | | 1.756e-05 | | | 0.90 +Comm | 0.00020234 | 0.00020234 | 0.00020234 | 0.0 | 7.19 +Output | 3.6436e-05 | 3.6436e-05 | 3.6436e-05 | 0.0 | 1.30 +Modify | 6.7542e-05 | 6.7542e-05 | 6.7542e-05 | 0.0 | 2.40 +Other | | 4.655e-05 | | | 1.65 -Nlocal: 256 ave 256 max 256 min +Nlocal: 256.000 ave 256 max 256 min Histogram: 1 0 0 0 0 0 0 0 0 0 -Nghost: 1431 ave 1431 max 1431 min +Nghost: 1431.00 ave 1431 max 1431 min Histogram: 1 0 0 0 0 0 0 0 0 0 -Neighs: 9688 ave 9688 max 9688 min +Neighs: 9688.00 ave 9688 max 9688 min Histogram: 1 0 0 0 0 0 0 0 0 0 Total # of neighbors = 9688 -Ave neighs/atom = 37.8438 +Ave neighs/atom = 37.843750 Neighbor list builds = 0 Dangerous builds not checked Setting up Verlet run ... Unit style : lj Current step : 31 Time step : 0.005 -Per MPI rank memory allocation (min/avg/max) = 2.63 | 2.63 | 2.63 Mbytes +Per MPI rank memory allocation (min/avg/max) = 2.630 | 2.630 | 2.630 Mbytes Step Temp E_pair E_mol TotEng Press 31 0.7494946 -5.7306417 0 -4.6107913 0.41043597 51 0.71349216 -5.6772387 0 -4.6111811 0.52117681 -Loop time of 0.00433063 on 1 procs for 20 steps with 256 atoms +Loop time of 0.00560916 on 1 procs for 20 steps with 256 atoms -Performance: 1995088.941 tau/day, 4618.261 timesteps/s -99.3% CPU use with 1 MPI tasks x no OpenMP threads +Performance: 1540338.414 tau/day, 3565.598 timesteps/s +99.2% CPU use with 1 MPI tasks x 1 OpenMP threads MPI task timing breakdown: Section | min time | avg time | max time |%varavg| %total --------------------------------------------------------------- -Pair | 0.0035121 | 0.0035121 | 0.0035121 | 0.0 | 81.10 -Neigh | 0.00050258 | 0.00050258 | 0.00050258 | 0.0 | 11.61 -Comm | 0.00019444 | 0.00019444 | 0.00019444 | 0.0 | 4.49 -Output | 1.2092e-05 | 1.2092e-05 | 1.2092e-05 | 0.0 | 0.28 -Modify | 7.2917e-05 | 7.2917e-05 | 7.2917e-05 | 0.0 | 1.68 -Other | | 3.647e-05 | | | 0.84 +Pair | 0.0044403 | 0.0044403 | 0.0044403 | 0.0 | 79.16 +Neigh | 0.00056186 | 0.00056186 | 0.00056186 | 0.0 | 10.02 +Comm | 0.00036797 | 0.00036797 | 0.00036797 | 0.0 | 6.56 +Output | 3.676e-05 | 3.676e-05 | 3.676e-05 | 0.0 | 0.66 +Modify | 0.00011282 | 0.00011282 | 0.00011282 | 0.0 | 2.01 +Other | | 8.943e-05 | | | 1.59 -Nlocal: 256 ave 256 max 256 min +Nlocal: 256.000 ave 256 max 256 min Histogram: 1 0 0 0 0 0 0 0 0 0 -Nghost: 1421 ave 1421 max 1421 min +Nghost: 1421.00 ave 1421 max 1421 min Histogram: 1 0 0 0 0 0 0 0 0 0 -Neighs: 9700 ave 9700 max 9700 min +Neighs: 9700.00 ave 9700 max 9700 min Histogram: 1 0 0 0 0 0 0 0 0 0 Total # of neighbors = 9700 -Ave neighs/atom = 37.8906 +Ave neighs/atom = 37.890625 Neighbor list builds = 1 Dangerous builds not checked Setting up Verlet run ... Unit style : lj Current step : 51 Time step : 0.005 -Per MPI rank memory allocation (min/avg/max) = 2.63 | 2.63 | 2.63 Mbytes +Per MPI rank memory allocation (min/avg/max) = 2.630 | 2.630 | 2.630 Mbytes Step Temp E_pair E_mol TotEng Press 51 0.71349216 -5.6772387 0 -4.6111811 0.52117681 61 0.78045421 -5.7781094 0 -4.6120011 0.093808941 -Loop time of 0.00196567 on 1 procs for 10 steps with 256 atoms +Loop time of 0.00373815 on 1 procs for 10 steps with 256 atoms -Performance: 2197727.285 tau/day, 5087.332 timesteps/s -99.7% CPU use with 1 MPI tasks x no OpenMP threads +Performance: 1155650.623 tau/day, 2675.117 timesteps/s +98.0% CPU use with 1 MPI tasks x 1 OpenMP threads MPI task timing breakdown: Section | min time | avg time | max time |%varavg| %total --------------------------------------------------------------- -Pair | 0.0018222 | 0.0018222 | 0.0018222 | 0.0 | 92.70 +Pair | 0.0030908 | 0.0030908 | 0.0030908 | 0.0 | 82.68 Neigh | 0 | 0 | 0 | 0.0 | 0.00 -Comm | 7.8285e-05 | 7.8285e-05 | 7.8285e-05 | 0.0 | 3.98 -Output | 1.0862e-05 | 1.0862e-05 | 1.0862e-05 | 0.0 | 0.55 -Modify | 3.6719e-05 | 3.6719e-05 | 3.6719e-05 | 0.0 | 1.87 -Other | | 1.764e-05 | | | 0.90 +Comm | 0.00038189 | 0.00038189 | 0.00038189 | 0.0 | 10.22 +Output | 4.1615e-05 | 4.1615e-05 | 4.1615e-05 | 0.0 | 1.11 +Modify | 0.00013851 | 0.00013851 | 0.00013851 | 0.0 | 3.71 +Other | | 8.533e-05 | | | 2.28 -Nlocal: 256 ave 256 max 256 min +Nlocal: 256.000 ave 256 max 256 min Histogram: 1 0 0 0 0 0 0 0 0 0 -Nghost: 1421 ave 1421 max 1421 min +Nghost: 1421.00 ave 1421 max 1421 min Histogram: 1 0 0 0 0 0 0 0 0 0 -Neighs: 9700 ave 9700 max 9700 min +Neighs: 9700.00 ave 9700 max 9700 min Histogram: 1 0 0 0 0 0 0 0 0 0 Total # of neighbors = 9700 -Ave neighs/atom = 37.8906 +Ave neighs/atom = 37.890625 Neighbor list builds = 0 Dangerous builds not checked Setting up Verlet run ... Unit style : lj Current step : 61 Time step : 0.005 -Per MPI rank memory allocation (min/avg/max) = 2.63 | 2.63 | 2.63 Mbytes +Per MPI rank memory allocation (min/avg/max) = 2.630 | 2.630 | 2.630 Mbytes Step Temp E_pair E_mol TotEng Press 61 0.78045421 -5.7781094 0 -4.6120011 0.093808941 81 0.77743907 -5.7735004 0 -4.6118971 0.090822641 -Loop time of 0.00430528 on 1 procs for 20 steps with 256 atoms +Loop time of 0.00612177 on 1 procs for 20 steps with 256 atoms -Performance: 2006838.581 tau/day, 4645.460 timesteps/s -99.8% CPU use with 1 MPI tasks x no OpenMP threads +Performance: 1411356.519 tau/day, 3267.029 timesteps/s +98.6% CPU use with 1 MPI tasks x 1 OpenMP threads MPI task timing breakdown: Section | min time | avg time | max time |%varavg| %total --------------------------------------------------------------- -Pair | 0.0034931 | 0.0034931 | 0.0034931 | 0.0 | 81.13 -Neigh | 0.00050437 | 0.00050437 | 0.00050437 | 0.0 | 11.72 -Comm | 0.0001868 | 0.0001868 | 0.0001868 | 0.0 | 4.34 -Output | 1.1699e-05 | 1.1699e-05 | 1.1699e-05 | 0.0 | 0.27 -Modify | 7.3308e-05 | 7.3308e-05 | 7.3308e-05 | 0.0 | 1.70 -Other | | 3.604e-05 | | | 0.84 +Pair | 0.0047211 | 0.0047211 | 0.0047211 | 0.0 | 77.12 +Neigh | 0.00083088 | 0.00083088 | 0.00083088 | 0.0 | 13.57 +Comm | 0.00032716 | 0.00032716 | 0.00032716 | 0.0 | 5.34 +Output | 3.9891e-05 | 3.9891e-05 | 3.9891e-05 | 0.0 | 0.65 +Modify | 0.00010926 | 0.00010926 | 0.00010926 | 0.0 | 1.78 +Other | | 9.346e-05 | | | 1.53 -Nlocal: 256 ave 256 max 256 min +Nlocal: 256.000 ave 256 max 256 min Histogram: 1 0 0 0 0 0 0 0 0 0 -Nghost: 1405 ave 1405 max 1405 min +Nghost: 1405.00 ave 1405 max 1405 min Histogram: 1 0 0 0 0 0 0 0 0 0 -Neighs: 9701 ave 9701 max 9701 min +Neighs: 9701.00 ave 9701 max 9701 min Histogram: 1 0 0 0 0 0 0 0 0 0 Total # of neighbors = 9701 -Ave neighs/atom = 37.8945 +Ave neighs/atom = 37.894531 Neighbor list builds = 1 Dangerous builds not checked Deleted 256 atoms, new total = 0 @@ -261,34 +264,34 @@ Setting up Verlet run ... Unit style : lj Current step : 81 Time step : 0.005 -Per MPI rank memory allocation (min/avg/max) = 2.63 | 2.63 | 2.63 Mbytes +Per MPI rank memory allocation (min/avg/max) = 2.630 | 2.630 | 2.630 Mbytes Step Temp E_pair E_mol TotEng Press 81 0.6239063 -5.5404291 0 -4.6082254 1.0394285 91 0.75393007 -5.7375259 0 -4.6110484 0.39357367 -Loop time of 0.00195843 on 1 procs for 10 steps with 256 atoms +Loop time of 0.00319065 on 1 procs for 10 steps with 256 atoms -Performance: 2205851.941 tau/day, 5106.139 timesteps/s -99.7% CPU use with 1 MPI tasks x no OpenMP threads +Performance: 1353954.393 tau/day, 3134.154 timesteps/s +99.2% CPU use with 1 MPI tasks x 1 OpenMP threads MPI task timing breakdown: Section | min time | avg time | max time |%varavg| %total --------------------------------------------------------------- -Pair | 0.0018143 | 0.0018143 | 0.0018143 | 0.0 | 92.64 +Pair | 0.0027828 | 0.0027828 | 0.0027828 | 0.0 | 87.22 Neigh | 0 | 0 | 0 | 0.0 | 0.00 -Comm | 7.8608e-05 | 7.8608e-05 | 7.8608e-05 | 0.0 | 4.01 -Output | 1.0786e-05 | 1.0786e-05 | 1.0786e-05 | 0.0 | 0.55 -Modify | 3.7106e-05 | 3.7106e-05 | 3.7106e-05 | 0.0 | 1.89 -Other | | 1.762e-05 | | | 0.90 +Comm | 0.00023286 | 0.00023286 | 0.00023286 | 0.0 | 7.30 +Output | 4.0459e-05 | 4.0459e-05 | 4.0459e-05 | 0.0 | 1.27 +Modify | 7.3576e-05 | 7.3576e-05 | 7.3576e-05 | 0.0 | 2.31 +Other | | 6.094e-05 | | | 1.91 -Nlocal: 256 ave 256 max 256 min +Nlocal: 256.000 ave 256 max 256 min Histogram: 1 0 0 0 0 0 0 0 0 0 -Nghost: 1431 ave 1431 max 1431 min +Nghost: 1431.00 ave 1431 max 1431 min Histogram: 1 0 0 0 0 0 0 0 0 0 -Neighs: 9705 ave 9705 max 9705 min +Neighs: 9705.00 ave 9705 max 9705 min Histogram: 1 0 0 0 0 0 0 0 0 0 Total # of neighbors = 9705 -Ave neighs/atom = 37.9102 +Ave neighs/atom = 37.910156 Neighbor list builds = 0 Dangerous builds not checked Total wall time: 0:00:00 diff --git a/examples/COUPLE/plugin/log.simple.plugin.4 b/examples/COUPLE/plugin/log.simple.plugin.4 index d76719dd7c..9b578d9919 100644 --- a/examples/COUPLE/plugin/log.simple.plugin.4 +++ b/examples/COUPLE/plugin/log.simple.plugin.4 @@ -1,9 +1,12 @@ -LAMMPS (18 Feb 2020) -Lattice spacing in x,y,z = 1.6796 1.6796 1.6796 -Created orthogonal box = (0 0 0) to (6.71838 6.71838 6.71838) +LAMMPS (31 Aug 2021) +OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (src/comm.cpp:98) + using 1 OpenMP thread(s) per MPI task +Lattice spacing in x,y,z = 1.6795962 1.6795962 1.6795962 +Created orthogonal box = (0.0000000 0.0000000 0.0000000) to (6.7183848 6.7183848 6.7183848) 1 by 1 by 2 MPI processor grid Created 256 atoms - create_atoms CPU = 0.000265157 secs + using lattice units in orthogonal box = (0.0000000 0.0000000 0.0000000) to (6.7183848 6.7183848 6.7183848) + create_atoms CPU = 0.003 seconds Neighbor list info ... update every 20 steps, delay 0 steps, check no max neighbors/atom: 2000, page size: 100000 @@ -14,7 +17,7 @@ Neighbor list info ... (1) pair lj/cut, perpetual attributes: half, newton on pair build: half/bin/atomonly/newton - stencil: half/bin/3d/newton + stencil: half/bin/3d bin: standard Setting up Verlet run ... Unit style : lj @@ -24,30 +27,30 @@ Per MPI rank memory allocation (min/avg/max) = 2.624 | 2.624 | 2.624 Mbytes Step Temp E_pair E_mol TotEng Press 0 1.44 -6.7733681 0 -4.6218056 -5.0244179 10 1.1298532 -6.3095502 0 -4.6213906 -2.6058175 -Loop time of 0.00115264 on 2 procs for 10 steps with 256 atoms +Loop time of 0.00330899 on 2 procs for 10 steps with 256 atoms -Performance: 3747912.946 tau/day, 8675.724 timesteps/s -94.5% CPU use with 2 MPI tasks x no OpenMP threads +Performance: 1305535.501 tau/day, 3022.073 timesteps/s +75.7% CPU use with 2 MPI tasks x 1 OpenMP threads MPI task timing breakdown: Section | min time | avg time | max time |%varavg| %total --------------------------------------------------------------- -Pair | 0.00074885 | 0.00075021 | 0.00075156 | 0.0 | 65.09 +Pair | 0.0013522 | 0.0013813 | 0.0014104 | 0.1 | 41.74 Neigh | 0 | 0 | 0 | 0.0 | 0.00 -Comm | 0.00031829 | 0.00031943 | 0.00032056 | 0.0 | 27.71 -Output | 9.306e-06 | 2.6673e-05 | 4.4041e-05 | 0.0 | 2.31 -Modify | 2.0684e-05 | 2.0891e-05 | 2.1098e-05 | 0.0 | 1.81 -Other | | 3.544e-05 | | | 3.07 +Comm | 0.00049139 | 0.00054241 | 0.00059342 | 0.0 | 16.39 +Output | 3.6986e-05 | 0.00056588 | 0.0010948 | 0.0 | 17.10 +Modify | 4.3909e-05 | 4.3924e-05 | 4.3939e-05 | 0.0 | 1.33 +Other | | 0.0007755 | | | 23.44 -Nlocal: 128 ave 128 max 128 min +Nlocal: 128.000 ave 128 max 128 min Histogram: 2 0 0 0 0 0 0 0 0 0 -Nghost: 1109 ave 1109 max 1109 min +Nghost: 1109.00 ave 1109 max 1109 min Histogram: 2 0 0 0 0 0 0 0 0 0 -Neighs: 4992 ave 4992 max 4992 min +Neighs: 4992.00 ave 4992 max 4992 min Histogram: 2 0 0 0 0 0 0 0 0 0 Total # of neighbors = 9984 -Ave neighs/atom = 39 +Ave neighs/atom = 39.000000 Neighbor list builds = 0 Dangerous builds not checked Setting up Verlet run ... @@ -58,30 +61,30 @@ Per MPI rank memory allocation (min/avg/max) = 2.624 | 2.624 | 2.624 Mbytes Step Temp E_pair E_mol TotEng Press 10 1.1298532 -6.3095502 0 -4.6213906 -2.6058175 20 0.6239063 -5.557644 0 -4.6254403 0.97451173 -Loop time of 0.00120443 on 2 procs for 10 steps with 256 atoms +Loop time of 0.00648485 on 2 procs for 10 steps with 256 atoms -Performance: 3586761.860 tau/day, 8302.689 timesteps/s -95.5% CPU use with 2 MPI tasks x no OpenMP threads +Performance: 666168.017 tau/day, 1542.056 timesteps/s +44.3% CPU use with 2 MPI tasks x 1 OpenMP threads MPI task timing breakdown: Section | min time | avg time | max time |%varavg| %total --------------------------------------------------------------- -Pair | 0.00087798 | 0.00091359 | 0.0009492 | 0.0 | 75.85 +Pair | 0.0022373 | 0.0024405 | 0.0026437 | 0.4 | 37.63 Neigh | 0 | 0 | 0 | 0.0 | 0.00 -Comm | 0.00016739 | 0.00020368 | 0.00023997 | 0.0 | 16.91 -Output | 1.0124e-05 | 3.0513e-05 | 5.0901e-05 | 0.0 | 2.53 -Modify | 1.89e-05 | 1.9812e-05 | 2.0725e-05 | 0.0 | 1.64 -Other | | 3.683e-05 | | | 3.06 +Comm | 0.0024446 | 0.0026464 | 0.0028481 | 0.4 | 40.81 +Output | 3.9069e-05 | 0.00059734 | 0.0011556 | 0.0 | 9.21 +Modify | 4.869e-05 | 4.912e-05 | 4.9551e-05 | 0.0 | 0.76 +Other | | 0.0007515 | | | 11.59 -Nlocal: 128 ave 134 max 122 min +Nlocal: 128.000 ave 134 max 122 min Histogram: 1 0 0 0 0 0 0 0 0 1 -Nghost: 1109 ave 1115 max 1103 min +Nghost: 1109.00 ave 1115 max 1103 min Histogram: 1 0 0 0 0 0 0 0 0 1 -Neighs: 4976 ave 5205 max 4747 min +Neighs: 4976.00 ave 5205 max 4747 min Histogram: 1 0 0 0 0 0 0 0 0 1 Total # of neighbors = 9952 -Ave neighs/atom = 38.875 +Ave neighs/atom = 38.875000 Neighbor list builds = 0 Dangerous builds not checked Setting up Verlet run ... @@ -92,34 +95,34 @@ Per MPI rank memory allocation (min/avg/max) = 2.624 | 2.624 | 2.624 Mbytes Step Temp E_pair E_mol TotEng Press 20 0.6239063 -5.5404291 0 -4.6082254 1.0394285 21 0.63845863 -5.5628733 0 -4.6089263 0.99398278 -Loop time of 0.000206062 on 2 procs for 1 steps with 256 atoms +Loop time of 0.00128072 on 2 procs for 1 steps with 256 atoms -Performance: 2096456.406 tau/day, 4852.908 timesteps/s -94.1% CPU use with 2 MPI tasks x no OpenMP threads +Performance: 337310.921 tau/day, 780.812 timesteps/s +60.2% CPU use with 2 MPI tasks x 1 OpenMP threads MPI task timing breakdown: Section | min time | avg time | max time |%varavg| %total --------------------------------------------------------------- -Pair | 0.00012947 | 0.00013524 | 0.00014101 | 0.0 | 65.63 +Pair | 0.00047351 | 0.00049064 | 0.00050777 | 0.0 | 38.31 Neigh | 0 | 0 | 0 | 0.0 | 0.00 -Comm | 1.858e-05 | 2.4113e-05 | 2.9647e-05 | 0.0 | 11.70 -Output | 8.699e-06 | 2.4204e-05 | 3.9708e-05 | 0.0 | 11.75 -Modify | 2.34e-06 | 2.3705e-06 | 2.401e-06 | 0.0 | 1.15 -Other | | 2.013e-05 | | | 9.77 +Comm | 7.6767e-05 | 9.3655e-05 | 0.00011054 | 0.0 | 7.31 +Output | 5.4217e-05 | 0.00026297 | 0.00047172 | 0.0 | 20.53 +Modify | 1.1554e-05 | 1.2026e-05 | 1.2498e-05 | 0.0 | 0.94 +Other | | 0.0004214 | | | 32.91 -Nlocal: 128 ave 135 max 121 min +Nlocal: 128.000 ave 135 max 121 min Histogram: 1 0 0 0 0 0 0 0 0 1 -Nghost: 1109 ave 1116 max 1102 min +Nghost: 1109.00 ave 1116 max 1102 min Histogram: 1 0 0 0 0 0 0 0 0 1 -Neighs: 4852.5 ave 5106 max 4599 min +Neighs: 4852.50 ave 5106 max 4599 min Histogram: 1 0 0 0 0 0 0 0 0 1 Total # of neighbors = 9705 -Ave neighs/atom = 37.9102 -Force on 1 atom via extract_atom: -18.109 -Force on 1 atom via extract_variable: -18.109 +Ave neighs/atom = 37.910156 Neighbor list builds = 0 Dangerous builds not checked +Force on 1 atom via extract_atom: -18.109 +Force on 1 atom via extract_variable: -18.109 Force on 1 atom via extract_atom: 26.9581 Force on 1 atom via extract_variable: 26.9581 Setting up Verlet run ... @@ -130,30 +133,30 @@ Per MPI rank memory allocation (min/avg/max) = 2.624 | 2.624 | 2.624 Mbytes Step Temp E_pair E_mol TotEng Press 21 0.63845863 -5.5628733 0 -4.6089263 0.99398278 31 0.7494946 -5.7306417 0 -4.6107913 0.41043597 -Loop time of 0.00119048 on 2 procs for 10 steps with 256 atoms +Loop time of 0.00784933 on 2 procs for 10 steps with 256 atoms -Performance: 3628802.105 tau/day, 8400.005 timesteps/s -98.0% CPU use with 2 MPI tasks x no OpenMP threads +Performance: 550365.761 tau/day, 1273.995 timesteps/s +59.6% CPU use with 2 MPI tasks x 1 OpenMP threads MPI task timing breakdown: Section | min time | avg time | max time |%varavg| %total --------------------------------------------------------------- -Pair | 0.00085276 | 0.00089699 | 0.00094123 | 0.0 | 75.35 +Pair | 0.0019235 | 0.0033403 | 0.0047572 | 2.5 | 42.56 Neigh | 0 | 0 | 0 | 0.0 | 0.00 -Comm | 0.00016896 | 0.00021444 | 0.00025992 | 0.0 | 18.01 -Output | 9.413e-06 | 2.5939e-05 | 4.2465e-05 | 0.0 | 2.18 -Modify | 1.8977e-05 | 2.0009e-05 | 2.1042e-05 | 0.0 | 1.68 -Other | | 3.31e-05 | | | 2.78 +Comm | 0.0016948 | 0.003118 | 0.0045411 | 2.5 | 39.72 +Output | 3.6445e-05 | 0.00064636 | 0.0012563 | 0.0 | 8.23 +Modify | 6.2842e-05 | 6.3209e-05 | 6.3577e-05 | 0.0 | 0.81 +Other | | 0.0006814 | | | 8.68 -Nlocal: 128 ave 135 max 121 min +Nlocal: 128.000 ave 135 max 121 min Histogram: 1 0 0 0 0 0 0 0 0 1 -Nghost: 1109 ave 1116 max 1102 min +Nghost: 1109.00 ave 1116 max 1102 min Histogram: 1 0 0 0 0 0 0 0 0 1 -Neighs: 4844 ave 5096 max 4592 min +Neighs: 4844.00 ave 5096 max 4592 min Histogram: 1 0 0 0 0 0 0 0 0 1 Total # of neighbors = 9688 -Ave neighs/atom = 37.8438 +Ave neighs/atom = 37.843750 Neighbor list builds = 0 Dangerous builds not checked Setting up Verlet run ... @@ -164,30 +167,30 @@ Per MPI rank memory allocation (min/avg/max) = 2.624 | 2.624 | 2.624 Mbytes Step Temp E_pair E_mol TotEng Press 31 0.7494946 -5.7306417 0 -4.6107913 0.41043597 51 0.71349216 -5.6772387 0 -4.6111811 0.52117681 -Loop time of 0.00252603 on 2 procs for 20 steps with 256 atoms +Loop time of 0.00696051 on 2 procs for 20 steps with 256 atoms -Performance: 3420382.192 tau/day, 7917.551 timesteps/s -99.2% CPU use with 2 MPI tasks x no OpenMP threads +Performance: 1241287.730 tau/day, 2873.351 timesteps/s +79.2% CPU use with 2 MPI tasks x 1 OpenMP threads MPI task timing breakdown: Section | min time | avg time | max time |%varavg| %total --------------------------------------------------------------- -Pair | 0.0016245 | 0.0017014 | 0.0017784 | 0.2 | 67.36 -Neigh | 0.00025359 | 0.0002563 | 0.00025901 | 0.0 | 10.15 -Comm | 0.00036863 | 0.00045124 | 0.00053385 | 0.0 | 17.86 -Output | 9.839e-06 | 2.8031e-05 | 4.6223e-05 | 0.0 | 1.11 -Modify | 3.7027e-05 | 3.9545e-05 | 4.2063e-05 | 0.0 | 1.57 -Other | | 4.948e-05 | | | 1.96 +Pair | 0.0028267 | 0.0036088 | 0.004391 | 1.3 | 51.85 +Neigh | 0.00040272 | 0.00040989 | 0.00041707 | 0.0 | 5.89 +Comm | 0.00081061 | 0.0015825 | 0.0023544 | 1.9 | 22.74 +Output | 3.6006e-05 | 0.00062106 | 0.0012061 | 0.0 | 8.92 +Modify | 6.8937e-05 | 7.1149e-05 | 7.336e-05 | 0.0 | 1.02 +Other | | 0.0006671 | | | 9.58 -Nlocal: 128 ave 132 max 124 min +Nlocal: 128.000 ave 132 max 124 min Histogram: 1 0 0 0 0 0 0 0 0 1 -Nghost: 1100 ave 1101 max 1099 min +Nghost: 1100.00 ave 1101 max 1099 min Histogram: 1 0 0 0 0 0 0 0 0 1 -Neighs: 4850 ave 4953 max 4747 min +Neighs: 4850.00 ave 4953 max 4747 min Histogram: 1 0 0 0 0 0 0 0 0 1 Total # of neighbors = 9700 -Ave neighs/atom = 37.8906 +Ave neighs/atom = 37.890625 Neighbor list builds = 1 Dangerous builds not checked Setting up Verlet run ... @@ -198,30 +201,30 @@ Per MPI rank memory allocation (min/avg/max) = 2.624 | 2.624 | 2.624 Mbytes Step Temp E_pair E_mol TotEng Press 51 0.71349216 -5.6772387 0 -4.6111811 0.52117681 61 0.78045421 -5.7781094 0 -4.6120011 0.093808941 -Loop time of 0.00115444 on 2 procs for 10 steps with 256 atoms +Loop time of 0.00155862 on 2 procs for 10 steps with 256 atoms -Performance: 3742065.976 tau/day, 8662.190 timesteps/s -96.5% CPU use with 2 MPI tasks x no OpenMP threads +Performance: 2771678.197 tau/day, 6415.922 timesteps/s +95.0% CPU use with 2 MPI tasks x 1 OpenMP threads MPI task timing breakdown: Section | min time | avg time | max time |%varavg| %total --------------------------------------------------------------- -Pair | 0.00087346 | 0.00089311 | 0.00091275 | 0.0 | 77.36 +Pair | 0.0012369 | 0.001266 | 0.001295 | 0.1 | 81.22 Neigh | 0 | 0 | 0 | 0.0 | 0.00 -Comm | 0.00016192 | 0.0001823 | 0.00020269 | 0.0 | 15.79 -Output | 9.49e-06 | 2.6234e-05 | 4.2978e-05 | 0.0 | 2.27 -Modify | 1.9095e-05 | 1.9843e-05 | 2.0591e-05 | 0.0 | 1.72 -Other | | 3.296e-05 | | | 2.85 +Comm | 0.00019462 | 0.00022315 | 0.00025169 | 0.0 | 14.32 +Output | 2.0217e-05 | 2.1945e-05 | 2.3673e-05 | 0.0 | 1.41 +Modify | 2.562e-05 | 2.5759e-05 | 2.5898e-05 | 0.0 | 1.65 +Other | | 2.181e-05 | | | 1.40 -Nlocal: 128 ave 132 max 124 min +Nlocal: 128.000 ave 132 max 124 min Histogram: 1 0 0 0 0 0 0 0 0 1 -Nghost: 1100 ave 1101 max 1099 min +Nghost: 1100.00 ave 1101 max 1099 min Histogram: 1 0 0 0 0 0 0 0 0 1 -Neighs: 4850 ave 4953 max 4747 min +Neighs: 4850.00 ave 4953 max 4747 min Histogram: 1 0 0 0 0 0 0 0 0 1 Total # of neighbors = 9700 -Ave neighs/atom = 37.8906 +Ave neighs/atom = 37.890625 Neighbor list builds = 0 Dangerous builds not checked Setting up Verlet run ... @@ -232,30 +235,30 @@ Per MPI rank memory allocation (min/avg/max) = 2.624 | 2.624 | 2.624 Mbytes Step Temp E_pair E_mol TotEng Press 61 0.78045421 -5.7781094 0 -4.6120011 0.093808941 81 0.77743907 -5.7735004 0 -4.6118971 0.090822641 -Loop time of 0.00244325 on 2 procs for 20 steps with 256 atoms +Loop time of 0.00351607 on 2 procs for 20 steps with 256 atoms -Performance: 3536279.919 tau/day, 8185.833 timesteps/s -99.0% CPU use with 2 MPI tasks x no OpenMP threads +Performance: 2457288.612 tau/day, 5688.168 timesteps/s +97.9% CPU use with 2 MPI tasks x 1 OpenMP threads MPI task timing breakdown: Section | min time | avg time | max time |%varavg| %total --------------------------------------------------------------- -Pair | 0.0016916 | 0.0017038 | 0.001716 | 0.0 | 69.73 -Neigh | 0.00025229 | 0.00025512 | 0.00025795 | 0.0 | 10.44 -Comm | 0.00035772 | 0.00036918 | 0.00038064 | 0.0 | 15.11 -Output | 1.0858e-05 | 2.7875e-05 | 4.4891e-05 | 0.0 | 1.14 -Modify | 3.817e-05 | 3.9325e-05 | 4.048e-05 | 0.0 | 1.61 -Other | | 4.796e-05 | | | 1.96 +Pair | 0.0023896 | 0.0024147 | 0.0024397 | 0.1 | 68.67 +Neigh | 0.00037331 | 0.00040456 | 0.0004358 | 0.0 | 11.51 +Comm | 0.00050571 | 0.00051343 | 0.00052116 | 0.0 | 14.60 +Output | 2.6424e-05 | 5.6547e-05 | 8.667e-05 | 0.0 | 1.61 +Modify | 5.0287e-05 | 5.1071e-05 | 5.1856e-05 | 0.0 | 1.45 +Other | | 7.58e-05 | | | 2.16 -Nlocal: 128 ave 128 max 128 min +Nlocal: 128.000 ave 128 max 128 min Histogram: 2 0 0 0 0 0 0 0 0 0 -Nghost: 1088.5 ave 1092 max 1085 min +Nghost: 1088.50 ave 1092 max 1085 min Histogram: 1 0 0 0 0 0 0 0 0 1 -Neighs: 4850.5 ave 4851 max 4850 min +Neighs: 4850.50 ave 4851 max 4850 min Histogram: 1 0 0 0 0 0 0 0 0 1 Total # of neighbors = 9701 -Ave neighs/atom = 37.8945 +Ave neighs/atom = 37.894531 Neighbor list builds = 1 Dangerous builds not checked Deleted 256 atoms, new total = 0 @@ -267,30 +270,30 @@ Per MPI rank memory allocation (min/avg/max) = 2.624 | 2.624 | 2.624 Mbytes Step Temp E_pair E_mol TotEng Press 81 0.6239063 -5.5404291 0 -4.6082254 1.0394285 91 0.75393007 -5.7375259 0 -4.6110484 0.39357367 -Loop time of 0.00118092 on 2 procs for 10 steps with 256 atoms +Loop time of 0.0109747 on 2 procs for 10 steps with 256 atoms -Performance: 3658158.625 tau/day, 8467.960 timesteps/s -98.6% CPU use with 2 MPI tasks x no OpenMP threads +Performance: 393631.731 tau/day, 911.185 timesteps/s +53.5% CPU use with 2 MPI tasks x 1 OpenMP threads MPI task timing breakdown: Section | min time | avg time | max time |%varavg| %total --------------------------------------------------------------- -Pair | 0.0008476 | 0.00089265 | 0.00093771 | 0.0 | 75.59 +Pair | 0.0012057 | 0.0012732 | 0.0013407 | 0.2 | 11.60 Neigh | 0 | 0 | 0 | 0.0 | 0.00 -Comm | 0.00016335 | 0.00020946 | 0.00025557 | 0.0 | 17.74 -Output | 8.87e-06 | 2.5733e-05 | 4.2595e-05 | 0.0 | 2.18 -Modify | 1.8755e-05 | 1.9814e-05 | 2.0872e-05 | 0.0 | 1.68 -Other | | 3.326e-05 | | | 2.82 +Comm | 0.00018882 | 0.00025686 | 0.00032489 | 0.0 | 2.34 +Output | 2.1943e-05 | 0.0047067 | 0.0093915 | 6.8 | 42.89 +Modify | 2.4614e-05 | 2.5439e-05 | 2.6264e-05 | 0.0 | 0.23 +Other | | 0.004712 | | | 42.94 -Nlocal: 128 ave 135 max 121 min +Nlocal: 128.000 ave 135 max 121 min Histogram: 1 0 0 0 0 0 0 0 0 1 -Nghost: 1109 ave 1116 max 1102 min +Nghost: 1109.00 ave 1116 max 1102 min Histogram: 1 0 0 0 0 0 0 0 0 1 -Neighs: 4852.5 ave 5106 max 4599 min +Neighs: 4852.50 ave 5106 max 4599 min Histogram: 1 0 0 0 0 0 0 0 0 1 Total # of neighbors = 9705 -Ave neighs/atom = 37.9102 +Ave neighs/atom = 37.910156 Neighbor list builds = 0 Dangerous builds not checked Total wall time: 0:00:00 diff --git a/examples/COUPLE/plugin/simple.c b/examples/COUPLE/plugin/simple.c index 541e4fd432..dc3934b40c 100644 --- a/examples/COUPLE/plugin/simple.c +++ b/examples/COUPLE/plugin/simple.c @@ -87,7 +87,7 @@ int main(int narg, char **arg) MPI_Abort(MPI_COMM_WORLD,1); } } - if (lammps == 1) plugin->open(0,NULL,comm_lammps,&lmp); + if (lammps == 1) lmp = plugin->open(0,NULL,comm_lammps,NULL); while (1) { if (me == 0) { @@ -139,7 +139,7 @@ int main(int narg, char **arg) cmds[0] = (char *)"run 10"; cmds[1] = (char *)"run 20"; - if (lammps == 1) plugin->commands_list(lmp,2,cmds); + if (lammps == 1) plugin->commands_list(lmp,2,(const char **)cmds); /* delete all atoms create_atoms() to create new ones with old coords, vels @@ -164,12 +164,13 @@ int main(int narg, char **arg) if (lammps == 1) { plugin->close(lmp); + MPI_Barrier(comm_lammps); + MPI_Comm_free(&comm_lammps); liblammpsplugin_release(plugin); } /* close down MPI */ - if (lammps == 1) MPI_Comm_free(&comm_lammps); MPI_Barrier(MPI_COMM_WORLD); MPI_Finalize(); }