-/* Copyright (c) 2007, 2008, 2009, 2010. The SimGrid Team.
+/* Copyright (c) 2007-2015. The SimGrid Team.
* All rights reserved. */
/* This program is free software; you can redistribute it and/or modify it
- * under the terms of the license (GNU LGPL) which comes with this package. */
-
-#include <stdint.h>
-#include <stdio.h>
-#include <stdlib.h>
+ * under the terms of the license (GNU LGPL) which comes with this package. */
#include "private.h"
#include "smpi_mpi_dt_private.h"
#include "mc/mc.h"
+#include "src/mc/mc_record.h"
+#include "xbt/replay.h"
#include "surf/surf.h"
-#include "simix/smx_private.h"
+#include "src/simix/smx_private.h"
#include "simgrid/sg_config.h"
+#include "src/mc/mc_replay.h"
+#include "src/msg/msg_private.h"
+#include <float.h> /* DBL_MAX */
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
XBT_LOG_NEW_DEFAULT_SUBCATEGORY(smpi_kernel, smpi,
"Logging specific to SMPI (kernel)");
typedef struct s_smpi_process_data {
- int index;
- int* argc;
- char*** argv;
+ double simulated;
+ int *argc;
+ char ***argv;
smx_rdv_t mailbox;
smx_rdv_t mailbox_small;
+ xbt_mutex_t mailboxes_mutex;
xbt_os_timer_t timer;
- double simulated;
MPI_Comm comm_self;
- void *data; /* user data */
- int initialized;
+ MPI_Comm comm_intra;
+ MPI_Comm* comm_world;
+ void *data; /* user data */
+ int index;
+ char state;
+ int sampling; /* inside an SMPI_SAMPLE_ block? */
+ char* instance_id;
+ int replaying; /* is the process replaying a trace */
+ xbt_bar_t finalization_barrier;
} s_smpi_process_data_t;
static smpi_process_data_t *process_data = NULL;
-static int process_count = 0;
+int process_count = 0;
+int smpi_universe_size = 0;
+int* index_to_process_data = NULL;
+extern double smpi_total_benched_time;
+xbt_os_timer_t global_timer;
+MPI_Comm MPI_COMM_WORLD = MPI_COMM_UNINITIALIZED;
-MPI_Comm MPI_COMM_WORLD = MPI_COMM_NULL;
-int MPI_UNIVERSE_SIZE;
-
-MPI_Errhandler* MPI_ERRORS_RETURN = NULL;
-MPI_Errhandler* MPI_ERRORS_ARE_FATAL = NULL;
-MPI_Errhandler* MPI_ERRHANDLER_NULL = NULL;
+MPI_Errhandler *MPI_ERRORS_RETURN = NULL;
+MPI_Errhandler *MPI_ERRORS_ARE_FATAL = NULL;
+MPI_Errhandler *MPI_ERRHANDLER_NULL = NULL;
#define MAILBOX_NAME_MAXLEN (5 + sizeof(int) * 2 + 1)
-static char* get_mailbox_name(char* str, int index) {
- snprintf(str, MAILBOX_NAME_MAXLEN, "SMPI-%0*x", (int)(sizeof(int) * 2), index);
+static char *get_mailbox_name(char *str, int index)
+{
+ snprintf(str, MAILBOX_NAME_MAXLEN, "SMPI-%0*x", (int) (sizeof(int) * 2),
+ index);
return str;
}
-static char* get_mailbox_name_small(char* str, int index) {
- snprintf(str, MAILBOX_NAME_MAXLEN, "small%0*x", (int)(sizeof(int) * 2), index);
+static char *get_mailbox_name_small(char *str, int index)
+{
+ snprintf(str, MAILBOX_NAME_MAXLEN, "small%0*x", (int) (sizeof(int) * 2),
+ index);
return str;
}
void smpi_process_init(int *argc, char ***argv)
{
- int index;
+ int index=-1;
smpi_process_data_t data;
smx_process_t proc;
- if(argc && argv) {
+ if (argc && argv) {
proc = SIMIX_process_self();
- index = atoi((*argv)[1]);
+ //FIXME: dirty cleanup method to avoid using msg cleanup functions on these processes when using MSG+SMPI
+ SIMIX_process_set_cleanup_function(proc, SIMIX_process_cleanup);
+ char* instance_id = (*argv)[1];
+ int rank = atoi((*argv)[2]);
+ index = smpi_process_index_of_smx_process(proc);
+
+ if(!index_to_process_data){
+ index_to_process_data=(int*)xbt_malloc(SIMIX_process_count()*sizeof(int));
+ }
+
+ if(smpi_privatize_global_variables){
+ /* Now using segment index of the process */
+ index = proc->segment_index;
+ /* Done at the process's creation */
+ SMPI_switch_data_segment(index);
+ }
+
+ MPI_Comm* temp_comm_world;
+ xbt_bar_t temp_bar;
+ smpi_deployment_register_process(instance_id, rank, index, &temp_comm_world ,&temp_bar);
data = smpi_process_remote_data(index);
- simcall_process_set_data(proc, data);
- if (*argc > 2) {
+ data->comm_world = temp_comm_world;
+ if(temp_bar != NULL) data->finalization_barrier = temp_bar;
+ data->index = index;
+ data->instance_id = instance_id;
+ data->replaying = 0;
+ //xbt_free(simcall_process_get_data(proc));
+
+ simdata_process_t simdata = simcall_process_get_data(proc);
+ simdata->data = data;
+
+ if (*argc > 3) {
free((*argv)[1]);
- memmove(&(*argv)[1], &(*argv)[2], sizeof(char *) * (*argc - 2));
+ memmove(&(*argv)[0], &(*argv)[2], sizeof(char *) * (*argc - 2));
(*argv)[(*argc) - 1] = NULL;
+ (*argv)[(*argc) - 2] = NULL;
}
- (*argc)--;
+ (*argc)-=2;
data->argc = argc;
data->argv = argv;
- simcall_rdv_set_receiver(data->mailbox_small, proc);// set the process attached to the mailbox
+ // set the process attached to the mailbox
+ simcall_rdv_set_receiver(data->mailbox_small, proc);
XBT_DEBUG("<%d> New process in the game: %p", index, proc);
+
}
+ if (smpi_process_data() == NULL)
+ xbt_die("smpi_process_data() returned NULL. You probably gave a NULL parameter to MPI_Init. Although it's required by MPI-2, this is currently not supported by SMPI.");
}
void smpi_process_destroy(void)
{
int index = smpi_process_index();
- process_data[index]->index=-100;
+ if(smpi_privatize_global_variables){
+ smpi_switch_data_segment(index);
+ }
+ process_data[index_to_process_data[index]]->state = SMPI_FINALIZED;
XBT_DEBUG("<%d> Process left the game", index);
}
*/
void smpi_process_finalize(void)
{
- // wait for all pending asynchronous comms to finish
- while (SIMIX_process_has_pending_comms(SIMIX_process_self())) {
- simcall_process_sleep(0.01);
- }
+ // This leads to an explosion of the search graph
+ // which cannot be reduced:
+ if(MC_is_active() || MC_record_replay_is_active())
+ return;
+
+ int index = smpi_process_index();
+ // wait for all pending asynchronous comms to finish
+ xbt_barrier_wait(process_data[index_to_process_data[index]]->finalization_barrier);
}
/**
*/
int smpi_process_finalized()
{
- return (smpi_process_index()==-100);
- // If finalized, this value has been set to -100;
+ int index = smpi_process_index();
+ if (index != MPI_UNDEFINED)
+ return (process_data[index_to_process_data[index]]->state == SMPI_FINALIZED);
+ else
+ return 0;
}
/**
*/
int smpi_process_initialized(void)
{
- int index = smpi_process_index();
- return((index != -100) && (index!=MPI_UNDEFINED) && (process_data[index]->initialized));
+ if (!index_to_process_data){
+ return false;
+ }
+ else{
+ int index = smpi_process_index();
+ return ( (index != MPI_UNDEFINED)
+ && (process_data[index_to_process_data[index]]->state == SMPI_INITIALIZED));
+ }
}
/**
void smpi_process_mark_as_initialized(void)
{
int index = smpi_process_index();
- if((index != -100)&& (index!=MPI_UNDEFINED))process_data[index]->initialized=1;
+ if ((index != MPI_UNDEFINED) && (process_data[index_to_process_data[index]]->state != SMPI_FINALIZED))
+ process_data[index_to_process_data[index]]->state = SMPI_INITIALIZED;
}
-
-#ifdef SMPI_F2C
-int smpi_process_argc(void) {
- smpi_process_data_t data = smpi_process_data();
-
- return data->argc ? *(data->argc) - 1 : 0;
+void smpi_process_set_replaying(int value){
+ int index = smpi_process_index();
+ if ((index != MPI_UNDEFINED) && (process_data[index_to_process_data[index]]->state != SMPI_FINALIZED))
+ process_data[index_to_process_data[index]]->replaying = value;
}
-int smpi_process_getarg(integer* index, char* dst, ftnlen len) {
- smpi_process_data_t data = smpi_process_data();
- char* arg;
- ftnlen i;
-
- if(!data->argc || !data->argv
- || *index < 1 || *index >= *(data->argc)) {
- return -1;
- }
- arg = (*data->argv)[*index];
- for(i = 0; i < len && arg[i] != '\0'; i++) {
- dst[i] = arg[i];
- }
- for(; i < len; i++) {
- dst[i] = ' ';
- }
- return 0;
+int smpi_process_get_replaying(){
+ int index = smpi_process_index();
+ if (index != MPI_UNDEFINED)
+ return process_data[index_to_process_data[index]]->replaying;
+ else return _xbt_replay_is_active();
}
-int smpi_global_size(void) {
- char* value = getenv("SMPI_GLOBAL_SIZE");
- if(!value) {
- fprintf(stderr, "Please set env var SMPI_GLOBAL_SIZE to expected number of processes.\n");
- xbt_abort();
- }
- return atoi(value);
+int smpi_global_size(void)
+{
+ char *value = getenv("SMPI_GLOBAL_SIZE");
+
+ if (!value) {
+ fprintf(stderr,
+ "Please set env var SMPI_GLOBAL_SIZE to expected number of processes.\n");
+ xbt_abort();
+ }
+ return atoi(value);
}
-#endif
smpi_process_data_t smpi_process_data(void)
{
- return SIMIX_process_self_get_data(SIMIX_process_self());
+ simdata_process_t simdata = SIMIX_process_self_get_data(SIMIX_process_self());
+ return simdata->data;
}
smpi_process_data_t smpi_process_remote_data(int index)
{
- return process_data[index];
+ return process_data[index_to_process_data[index]];
}
void smpi_process_set_user_data(void *data)
process_data->data = data;
}
-void* smpi_process_get_user_data(){
+void *smpi_process_get_user_data()
+{
smpi_process_data_t process_data = smpi_process_data();
return process_data->data;
}
{
smpi_process_data_t data = smpi_process_data();
//return -1 if not initialized
- return data? data->index : MPI_UNDEFINED;
+ return data ? data->index : MPI_UNDEFINED;
}
-smx_rdv_t smpi_process_mailbox(void) {
+MPI_Comm smpi_process_comm_world(void)
+{
smpi_process_data_t data = smpi_process_data();
+ //return MPI_COMM_NULL if not initialized
+ return data ? *data->comm_world : MPI_COMM_NULL;
+}
+smx_rdv_t smpi_process_mailbox(void)
+{
+ smpi_process_data_t data = smpi_process_data();
return data->mailbox;
}
-smx_rdv_t smpi_process_mailbox_small(void) {
+smx_rdv_t smpi_process_mailbox_small(void)
+{
smpi_process_data_t data = smpi_process_data();
-
return data->mailbox_small;
}
-smx_rdv_t smpi_process_remote_mailbox(int index) {
- smpi_process_data_t data = smpi_process_remote_data(index);
+xbt_mutex_t smpi_process_mailboxes_mutex(void)
+{
+ smpi_process_data_t data = smpi_process_data();
+ return data->mailboxes_mutex;
+}
+smx_rdv_t smpi_process_remote_mailbox(int index)
+{
+ smpi_process_data_t data = smpi_process_remote_data(index);
return data->mailbox;
}
-smx_rdv_t smpi_process_remote_mailbox_small(int index) {
+smx_rdv_t smpi_process_remote_mailbox_small(int index)
+{
smpi_process_data_t data = smpi_process_remote_data(index);
-
return data->mailbox_small;
}
+xbt_mutex_t smpi_process_remote_mailboxes_mutex(int index)
+{
+ smpi_process_data_t data = smpi_process_remote_data(index);
+ return data->mailboxes_mutex;
+}
+
xbt_os_timer_t smpi_process_timer(void)
{
smpi_process_data_t data = smpi_process_data();
-
return data->timer;
}
-void smpi_process_simulated_start(void) {
+void smpi_process_simulated_start(void)
+{
smpi_process_data_t data = smpi_process_data();
-
data->simulated = SIMIX_get_clock();
}
-double smpi_process_simulated_elapsed(void) {
+double smpi_process_simulated_elapsed(void)
+{
smpi_process_data_t data = smpi_process_data();
-
return SIMIX_get_clock() - data->simulated;
}
-MPI_Comm smpi_process_comm_self(void) {
+MPI_Comm smpi_process_comm_self(void)
+{
smpi_process_data_t data = smpi_process_data();
+ if(data->comm_self==MPI_COMM_NULL){
+ MPI_Group group = smpi_group_new(1);
+ data->comm_self = smpi_comm_new(group, NULL);
+ smpi_group_set_mapping(group, smpi_process_index(), 0);
+ }
return data->comm_self;
}
-void print_request(const char *message, MPI_Request request) {
- XBT_DEBUG("%s request %p [buf = %p, size = %zu, src = %d, dst = %d, tag = %d, flags = %x]",
- message, request, request->buf, request->size,
- request->src, request->dst, request->tag, request->flags);
+MPI_Comm smpi_process_get_comm_intra(void)
+{
+ smpi_process_data_t data = smpi_process_data();
+ return data->comm_intra;
+}
+
+void smpi_process_set_comm_intra(MPI_Comm comm)
+{
+ smpi_process_data_t data = smpi_process_data();
+ data->comm_intra = comm;
+}
+
+void smpi_process_set_sampling(int s)
+{
+ smpi_process_data_t data = smpi_process_data();
+ data->sampling = s;
}
-static void SMPI_comm_copy_buffer_callback(smx_action_t comm, void* buff, size_t buff_size)
+int smpi_process_get_sampling(void)
+{
+ smpi_process_data_t data = smpi_process_data();
+ return data->sampling;
+}
+
+
+void print_request(const char *message, MPI_Request request)
+{
+ XBT_VERB
+ ("%s request %p [buf = %p, size = %zu, src = %d, dst = %d, tag = %d, flags = %x]",
+ message, request, request->buf, request->size, request->src,
+ request->dst, request->tag, request->flags);
+}
+
+void smpi_comm_copy_buffer_callback(smx_synchro_t comm,
+ void *buff, size_t buff_size)
{
XBT_DEBUG("Copy the data over");
- memcpy(comm->comm.dst_buff, buff, buff_size);
- if (comm->comm.detached) { // if this is a detached send, the source buffer was duplicated by SMPI sender to make the original buffer available to the application ASAP
+ void* tmpbuff=buff;
+
+ if((smpi_privatize_global_variables)
+ && ((char*)buff >= smpi_start_data_exe)
+ && ((char*)buff < smpi_start_data_exe + smpi_size_data_exe )
+ ){
+ XBT_DEBUG("Privatization : We are copying from a zone inside global memory... Saving data to temp buffer !");
+ smpi_switch_data_segment(((smpi_process_data_t)(((simdata_process_t)SIMIX_process_get_data(comm->comm.src_proc))->data))->index);
+ tmpbuff = (void*)xbt_malloc(buff_size);
+ memcpy(tmpbuff, buff, buff_size);
+ }
+
+
+ if((smpi_privatize_global_variables)
+ && ((char*)comm->comm.dst_buff >= smpi_start_data_exe)
+ && ((char*)comm->comm.dst_buff < smpi_start_data_exe + smpi_size_data_exe )
+ ){
+ XBT_DEBUG("Privatization : We are copying to a zone inside global memory - Switch data segment");
+ smpi_switch_data_segment(((smpi_process_data_t)(((simdata_process_t)SIMIX_process_get_data(comm->comm.dst_proc))->data))->index);
+ }
+
+
+ memcpy(comm->comm.dst_buff, tmpbuff, buff_size);
+ if (comm->comm.detached) {
+ // if this is a detached send, the source buffer was duplicated by SMPI
+ // sender to make the original buffer available to the application ASAP
xbt_free(buff);
//It seems that the request is used after the call there this should
//be free somewhereelse but where???
//xbt_free(comm->comm.src_data);// inside SMPI the request is keep
- //inside the user data and should be free
+ //inside the user data and should be free
comm->comm.src_buff = NULL;
}
+
+ if(tmpbuff!=buff)xbt_free(tmpbuff);
+
+}
+
+
+void smpi_comm_null_copy_buffer_callback(smx_synchro_t comm,
+ void *buff, size_t buff_size)
+{
+ return;
+}
+
+static void smpi_check_options(){
+ //check correctness of MPI parameters
+
+ xbt_assert(sg_cfg_get_int("smpi/async_small_thresh") <=
+ sg_cfg_get_int("smpi/send_is_detached_thres"));
+
+ if (sg_cfg_is_default_value("smpi/running_power")) {
+ XBT_INFO("You did not set the power of the host running the simulation. "
+ "The timings will certainly not be accurate. "
+ "Use the option \"--cfg=smpi/running_power:<flops>\" to set its value."
+ "Check http://simgrid.org/simgrid/latest/doc/options.html#options_smpi_bench for more information.");
+ }
+}
+
+int smpi_enabled(void) {
+ return process_data != NULL;
}
void smpi_global_init(void)
int i;
MPI_Group group;
char name[MAILBOX_NAME_MAXLEN];
+ int smpirun=0;
- SIMIX_comm_set_copy_data_callback(&SMPI_comm_copy_buffer_callback);
- process_count = SIMIX_process_count();
- process_data = xbt_new(smpi_process_data_t, process_count);
+ if (!MC_is_active()) {
+ global_timer = xbt_os_timer_new();
+ xbt_os_walltimer_start(global_timer);
+ }
+ if (process_count == 0){
+ process_count = SIMIX_process_count();
+ smpirun=1;
+ }
+ smpi_universe_size = process_count;
+ process_data = xbt_new0(smpi_process_data_t, process_count);
for (i = 0; i < process_count; i++) {
process_data[i] = xbt_new(s_smpi_process_data_t, 1);
- process_data[i]->index = i;
- process_data[i]->argc = NULL;
- process_data[i]->argv = NULL;
+ //process_data[i]->index = i;
+ process_data[i]->argc = NULL;
+ process_data[i]->argv = NULL;
process_data[i]->mailbox = simcall_rdv_create(get_mailbox_name(name, i));
- process_data[i]->mailbox_small = simcall_rdv_create(get_mailbox_name_small(name, i));
- process_data[i]->timer = xbt_os_timer_new();
- if(MC_is_active())
+ process_data[i]->mailbox_small =
+ simcall_rdv_create(get_mailbox_name_small(name, i));
+ process_data[i]->mailboxes_mutex = xbt_mutex_init();
+ process_data[i]->timer = xbt_os_timer_new();
+ if (MC_is_active())
MC_ignore_heap(process_data[i]->timer, xbt_os_timer_size());
- group = smpi_group_new(1);
- process_data[i]->comm_self = smpi_comm_new(group);
- process_data[i]->initialized =0;
-
- smpi_group_set_mapping(group, i, 0);
+ process_data[i]->comm_self = MPI_COMM_NULL;
+ process_data[i]->comm_intra = MPI_COMM_NULL;
+ process_data[i]->comm_world = NULL;
+ process_data[i]->state = SMPI_UNINITIALIZED;
+ process_data[i]->sampling = 0;
+ process_data[i]->finalization_barrier = NULL;
}
- group = smpi_group_new(process_count);
- MPI_COMM_WORLD = smpi_comm_new(group);
- MPI_UNIVERSE_SIZE = smpi_comm_size(MPI_COMM_WORLD);
- for (i = 0; i < process_count; i++) {
- smpi_group_set_mapping(group, i, i);
+ //if the process was launched through smpirun script
+ //we generate a global mpi_comm_world
+ //if not, we let MPI_COMM_NULL, and the comm world
+ //will be private to each mpi instance
+ if(smpirun){
+ group = smpi_group_new(process_count);
+ MPI_COMM_WORLD = smpi_comm_new(group, NULL);
+ MPI_Attr_put(MPI_COMM_WORLD, MPI_UNIVERSE_SIZE, (void *)(MPI_Aint)process_count);
+ xbt_bar_t bar=xbt_barrier_init(process_count);
+
+ for (i = 0; i < process_count; i++) {
+ smpi_group_set_mapping(group, i, i);
+ process_data[i]->finalization_barrier = bar;
+ }
}
-
- //check correctness of MPI parameters
-
- xbt_assert(sg_cfg_get_int("smpi/async_small_thres")<=sg_cfg_get_int("smpi/send_is_detached_thres"));
}
void smpi_global_destroy(void)
int i;
smpi_bench_destroy();
- while(smpi_group_unuse(smpi_comm_group(MPI_COMM_WORLD))>0);
- xbt_free(MPI_COMM_WORLD);
+ if (MPI_COMM_WORLD != MPI_COMM_UNINITIALIZED){
+ while (smpi_group_unuse(smpi_comm_group(MPI_COMM_WORLD)) > 0);
+ xbt_free(MPI_COMM_WORLD);
+ xbt_barrier_destroy(process_data[0]->finalization_barrier);
+ }else{
+ smpi_deployment_cleanup_instances();
+ }
MPI_COMM_WORLD = MPI_COMM_NULL;
for (i = 0; i < count; i++) {
- smpi_group_unuse(smpi_comm_group(process_data[i]->comm_self));
- smpi_comm_destroy(process_data[i]->comm_self);
+ if(process_data[i]->comm_self!=MPI_COMM_NULL){
+ smpi_group_unuse(smpi_comm_group(process_data[i]->comm_self));
+ smpi_comm_destroy(process_data[i]->comm_self);
+ }
+ if(process_data[i]->comm_intra!=MPI_COMM_NULL){
+ smpi_group_unuse(smpi_comm_group(process_data[i]->comm_intra));
+ smpi_comm_destroy(process_data[i]->comm_intra);
+ }
xbt_os_timer_free(process_data[i]->timer);
simcall_rdv_destroy(process_data[i]->mailbox);
simcall_rdv_destroy(process_data[i]->mailbox_small);
+ xbt_mutex_destroy(process_data[i]->mailboxes_mutex);
xbt_free(process_data[i]);
}
xbt_free(process_data);
process_data = NULL;
+ xbt_free(index_to_process_data);
+ if(smpi_privatize_global_variables)
+ smpi_destroy_global_memory_segments();
smpi_free_static();
}
-/* Fortran specific stuff */
-/* With smpicc, the following weak symbols are used */
-/* With smpiff, the following weak symbols are replaced by those in libf2c */
-int __attribute__((weak)) xargc;
-char** __attribute__((weak)) xargv;
-
#ifndef WIN32
-void __attribute__((weak)) user_main_(){
+void __attribute__ ((weak)) user_main_()
+{
xbt_die("Should not be in this smpi_simulated_main");
return;
}
-int __attribute__((weak)) smpi_simulated_main_(int argc, char** argv) {
+
+int __attribute__ ((weak)) smpi_simulated_main_(int argc, char **argv)
+{
smpi_process_init(&argc, &argv);
user_main_();
- //xbt_die("Should not be in this smpi_simulated_main");
return 0;
}
-int __attribute__((weak)) main(int argc, char** argv) {
- return smpi_main(smpi_simulated_main_,argc,argv);
+int __attribute__ ((weak)) main(int argc, char **argv)
+{
+ return smpi_main(smpi_simulated_main_, argc, argv);
}
-int __attribute__((weak)) MAIN__(){
- return smpi_main(smpi_simulated_main_,xargc, xargv);
-};
#endif
-int smpi_main(int (*realmain) (int argc, char *argv[]),int argc, char *argv[])
-{
- srand(SMPI_RAND_SEED);
-
- if(getenv("SMPI_PRETEND_CC") != NULL) {
- /* Hack to ensure that smpicc can pretend to be a simple compiler. Particularly handy to pass it to the configuration tools */
- return 0;
- }
+static void smpi_init_logs(){
/* Connect log categories. See xbt/log.c */
- XBT_LOG_CONNECT(smpi); /* Keep this line as soon as possible in this
- function: xbt_log_appender_file.c depends on it
- DO NOT connect this in XBT or so, or it will be
- useless to xbt_log_appender_file.c */
-#ifdef HAVE_TRACING
+ XBT_LOG_CONNECT(smpi); /* Keep this line as soon as possible in this
+ function: xbt_log_appender_file.c depends on it
+ DO NOT connect this in XBT or so, or it will be
+ useless to xbt_log_appender_file.c */
XBT_LOG_CONNECT(instr_smpi);
-#endif
XBT_LOG_CONNECT(smpi_base);
XBT_LOG_CONNECT(smpi_bench);
XBT_LOG_CONNECT(smpi_coll);
+ XBT_LOG_CONNECT(smpi_colls);
XBT_LOG_CONNECT(smpi_comm);
+ XBT_LOG_CONNECT(smpi_dvfs);
XBT_LOG_CONNECT(smpi_group);
XBT_LOG_CONNECT(smpi_kernel);
XBT_LOG_CONNECT(smpi_mpi);
XBT_LOG_CONNECT(smpi_mpi_dt);
XBT_LOG_CONNECT(smpi_pmpi);
XBT_LOG_CONNECT(smpi_replay);
- XBT_LOG_CONNECT(smpi_colls);
+ XBT_LOG_CONNECT(smpi_rma);
+
+}
+
+
+static void smpi_init_options(){
+ int gather_id = find_coll_description(mpi_coll_gather_description,
+ sg_cfg_get_string("smpi/gather"),"gather");
+ mpi_coll_gather_fun = (int (*)(void *, int, MPI_Datatype,
+ void *, int, MPI_Datatype, int, MPI_Comm))
+ mpi_coll_gather_description[gather_id].coll;
+
+ int allgather_id = find_coll_description(mpi_coll_allgather_description,
+ sg_cfg_get_string("smpi/allgather"),"allgather");
+ mpi_coll_allgather_fun = (int (*)(void *, int, MPI_Datatype,
+ void *, int, MPI_Datatype, MPI_Comm))
+ mpi_coll_allgather_description[allgather_id].coll;
+
+ int allgatherv_id = find_coll_description(mpi_coll_allgatherv_description,
+ sg_cfg_get_string("smpi/allgatherv"),"allgatherv");
+ mpi_coll_allgatherv_fun = (int (*)(void *, int, MPI_Datatype, void *, int *,
+ int *, MPI_Datatype, MPI_Comm))
+ mpi_coll_allgatherv_description[allgatherv_id].coll;
+
+ int allreduce_id = find_coll_description(mpi_coll_allreduce_description,
+ sg_cfg_get_string("smpi/allreduce"),"allreduce");
+ mpi_coll_allreduce_fun = (int (*)(void *sbuf, void *rbuf, int rcount,
+ MPI_Datatype dtype, MPI_Op op,
+ MPI_Comm comm))
+ mpi_coll_allreduce_description[allreduce_id].coll;
+
+ int alltoall_id = find_coll_description(mpi_coll_alltoall_description,
+ sg_cfg_get_string("smpi/alltoall"),"alltoall");
+ mpi_coll_alltoall_fun = (int (*)(void *, int, MPI_Datatype,
+ void *, int, MPI_Datatype, MPI_Comm))
+ mpi_coll_alltoall_description[alltoall_id].coll;
+
+ int alltoallv_id = find_coll_description(mpi_coll_alltoallv_description,
+ sg_cfg_get_string("smpi/alltoallv"),"alltoallv");
+ mpi_coll_alltoallv_fun = (int (*)(void *, int *, int *, MPI_Datatype,
+ void *, int *, int *, MPI_Datatype,
+ MPI_Comm))
+ mpi_coll_alltoallv_description[alltoallv_id].coll;
+
+ int bcast_id = find_coll_description(mpi_coll_bcast_description,
+ sg_cfg_get_string("smpi/bcast"),"bcast");
+ mpi_coll_bcast_fun = (int (*)(void *buf, int count, MPI_Datatype datatype,
+ int root, MPI_Comm com))
+ mpi_coll_bcast_description[bcast_id].coll;
+
+ int reduce_id = find_coll_description(mpi_coll_reduce_description,
+ sg_cfg_get_string("smpi/reduce"),"reduce");
+ mpi_coll_reduce_fun = (int (*)(void *buf, void *rbuf, int count,
+ MPI_Datatype datatype, MPI_Op op,
+ int root, MPI_Comm comm))
+ mpi_coll_reduce_description[reduce_id].coll;
+
+ int reduce_scatter_id =
+ find_coll_description(mpi_coll_reduce_scatter_description,
+ sg_cfg_get_string("smpi/reduce_scatter"),"reduce_scatter");
+ mpi_coll_reduce_scatter_fun = (int (*)(void *sbuf, void *rbuf, int *rcounts,
+ MPI_Datatype dtype, MPI_Op op,
+ MPI_Comm comm))
+ mpi_coll_reduce_scatter_description[reduce_scatter_id].coll;
+
+ int scatter_id = find_coll_description(mpi_coll_scatter_description,
+ sg_cfg_get_string("smpi/scatter"),"scatter");
+ mpi_coll_scatter_fun = (int (*)(void *sendbuf, int sendcount,
+ MPI_Datatype sendtype, void *recvbuf,
+ int recvcount, MPI_Datatype recvtype,
+ int root, MPI_Comm comm))
+ mpi_coll_scatter_description[scatter_id].coll;
+
+ int barrier_id = find_coll_description(mpi_coll_barrier_description,
+ sg_cfg_get_string("smpi/barrier"),"barrier");
+ mpi_coll_barrier_fun = (int (*)(MPI_Comm comm))
+ mpi_coll_barrier_description[barrier_id].coll;
+
+ smpi_cpu_threshold = sg_cfg_get_double("smpi/cpu_threshold");
+ smpi_running_power = sg_cfg_get_double("smpi/running_power");
+ smpi_privatize_global_variables = sg_cfg_get_boolean("smpi/privatize_global_variables");
+ if (smpi_cpu_threshold < 0)
+ smpi_cpu_threshold = DBL_MAX;
+
+}
+
+int smpi_main(int (*realmain) (int argc, char *argv[]), int argc, char *argv[])
+{
+ srand(SMPI_RAND_SEED);
+
+ if (getenv("SMPI_PRETEND_CC") != NULL) {
+ /* Hack to ensure that smpicc can pretend to be a simple
+ * compiler. Particularly handy to pass it to the configuration tools */
+ return 0;
+ }
+
+ smpi_init_logs();
-#ifdef HAVE_TRACING
TRACE_global_init(&argc, argv);
TRACE_add_start_function(TRACE_smpi_alloc);
TRACE_add_end_function(TRACE_smpi_release);
-#endif
SIMIX_global_init(&argc, argv);
+ MSG_init(&argc,argv);
-#ifdef HAVE_TRACING
- TRACE_start();
-#endif
+ SMPI_switch_data_segment = smpi_switch_data_segment;
+
+ smpi_init_options();
// parse the platform file: get the host list
SIMIX_create_environment(argv[1]);
-
+ SIMIX_comm_set_copy_data_callback(&smpi_comm_copy_buffer_callback);
SIMIX_function_register_default(realmain);
SIMIX_launch_application(argv[2]);
- int gather_id = find_coll_description(mpi_coll_gather_description,
- sg_cfg_get_string("smpi/gather"));
- mpi_coll_gather_fun = (int (*)(void *, int, MPI_Datatype,
- void*, int, MPI_Datatype, int, MPI_Comm))
- mpi_coll_gather_description[gather_id].coll;
-
- int allgather_id = find_coll_description(mpi_coll_allgather_description,
- sg_cfg_get_string("smpi/allgather"));
- mpi_coll_allgather_fun = (int (*)(void *, int, MPI_Datatype,
- void*, int, MPI_Datatype, MPI_Comm))
- mpi_coll_allgather_description[allgather_id].coll;
-
- int allgatherv_id = find_coll_description(mpi_coll_allgatherv_description,
- sg_cfg_get_string("smpi/allgatherv"));
- mpi_coll_allgatherv_fun = (int (*)(void *, int, MPI_Datatype,
- void*, int*, int*, MPI_Datatype, MPI_Comm))
- mpi_coll_allgatherv_description[allgatherv_id].coll;
-
- int allreduce_id = find_coll_description(mpi_coll_allreduce_description,
- sg_cfg_get_string("smpi/allreduce"));
- mpi_coll_allreduce_fun = (int (*)(void *sbuf, void *rbuf, int rcount, \
- MPI_Datatype dtype, MPI_Op op, MPI_Comm comm))
- mpi_coll_allreduce_description[allreduce_id].coll;
-
- int alltoall_id = find_coll_description(mpi_coll_alltoall_description,
- sg_cfg_get_string("smpi/alltoall"));
- mpi_coll_alltoall_fun = (int (*)(void *, int, MPI_Datatype,
- void*, int, MPI_Datatype, MPI_Comm))
- mpi_coll_alltoall_description[alltoall_id].coll;
-
- int alltoallv_id = find_coll_description(mpi_coll_alltoallv_description,
- sg_cfg_get_string("smpi/alltoallv"));
- mpi_coll_alltoallv_fun = (int (*)(void *, int*, int*, MPI_Datatype,
- void*, int*, int*, MPI_Datatype, MPI_Comm))
- mpi_coll_alltoallv_description[alltoallv_id].coll;
-
- int bcast_id = find_coll_description(mpi_coll_bcast_description,
- sg_cfg_get_string("smpi/bcast"));
- mpi_coll_bcast_fun = (int (*)(void *buf, int count, MPI_Datatype datatype, \
- int root, MPI_Comm com))
- mpi_coll_bcast_description[bcast_id].coll;
-
- int reduce_id = find_coll_description(mpi_coll_reduce_description,
- sg_cfg_get_string("smpi/reduce"));
- mpi_coll_reduce_fun = (int (*)(void *buf, void *rbuf, int count, MPI_Datatype datatype, \
- MPI_Op op, int root, MPI_Comm comm))
- mpi_coll_reduce_description[reduce_id].coll;
-
- int reduce_scatter_id = find_coll_description(mpi_coll_reduce_scatter_description,
- sg_cfg_get_string("smpi/reduce_scatter"));
- mpi_coll_reduce_scatter_fun = (int (*)(void *sbuf, void *rbuf, int *rcounts,\
- MPI_Datatype dtype,MPI_Op op,MPI_Comm comm))
- mpi_coll_reduce_scatter_description[reduce_scatter_id].coll;
-
- int scatter_id = find_coll_description(mpi_coll_scatter_description,
- sg_cfg_get_string("smpi/scatter"));
- mpi_coll_scatter_fun = (int (*)(void *sendbuf, int sendcount, MPI_Datatype sendtype,\
- void *recvbuf, int recvcount, MPI_Datatype recvtype,\
- int root, MPI_Comm comm))
- mpi_coll_scatter_description[scatter_id].coll;
-
- int barrier_id = find_coll_description(mpi_coll_barrier_description,
- sg_cfg_get_string("smpi/barrier"));
- mpi_coll_barrier_fun = (int (*)(MPI_Comm comm))
- mpi_coll_barrier_description[barrier_id].coll;
-
smpi_global_init();
+ smpi_check_options();
+
+ if(smpi_privatize_global_variables)
+ smpi_initialize_global_memory_segments();
+
/* Clean IO before the run */
fflush(stdout);
fflush(stderr);
- if (MC_is_active())
- MC_do_the_modelcheck_for_real();
- else
+ if (MC_is_active()) {
+ MC_run();
+ } else {
+
SIMIX_run();
- if (sg_cfg_get_boolean("smpi/display_timing"))
- XBT_INFO("Simulation time: %g seconds.", SIMIX_get_clock());
+ xbt_os_walltimer_stop(global_timer);
+ if (sg_cfg_get_boolean("smpi/display_timing")){
+ double global_time = xbt_os_timer_elapsed(global_timer);
+ XBT_INFO("Simulated time: %g seconds. \n\n"
+ "The simulation took %g seconds (after parsing and platform setup)\n"
+ "%g seconds were actual computation of the application"
+ , SIMIX_get_clock(), global_time , smpi_total_benched_time);
+
+ if (smpi_total_benched_time/global_time>=0.75)
+ XBT_INFO("More than 75%% of the time was spent inside the application code.\n"
+ "You may want to use sampling functions or trace replay to reduce this.");
+ }
+ }
smpi_global_destroy();
-#ifdef HAVE_TRACING
TRACE_end();
-#endif
return 0;
}
+
+// This function can be called from extern file, to initialize logs, options, and processes of smpi
+// without the need of smpirun
+void SMPI_init(){
+ smpi_init_logs();
+ smpi_init_options();
+ smpi_global_init();
+ smpi_check_options();
+ if (TRACE_is_enabled() && TRACE_is_configured())
+ TRACE_smpi_alloc();
+ if(smpi_privatize_global_variables)
+ smpi_initialize_global_memory_segments();
+}
+
+void SMPI_finalize(){
+ smpi_global_destroy();
+}