XBT_LOG_EXTERNAL_DEFAULT_CATEGORY(smpi_pmpi);
#define CHECK_RMA\
- CHECK_BUFFER(1, origin_addr, origin_count)\
CHECK_COUNT(2, origin_count)\
CHECK_TYPE(3, origin_datatype)\
- CHECK_PROC(4, target_rank)\
- CHECK_NEGATIVE(4, MPI_ERR_RANK, target_rank)\
+ CHECK_BUFFER(1, origin_addr, origin_count, origin_datatype)\
+ CHECK_PROC_RMA(4, target_rank, win)\
CHECK_COUNT(6, target_count)\
CHECK_TYPE(7, target_datatype)
+
+#define CHECK_TARGET_DISP(num) \
+ if (not win->dynamic()) \
+ CHECK_NEGATIVE((num), MPI_ERR_RMA_RANGE, target_disp)
+
/* PMPI User level calls */
int PMPI_Win_create( void *base, MPI_Aint size, int disp_unit, MPI_Info info, MPI_Comm comm, MPI_Win *win){
CHECK_COMM(5)
CHECK_NEGATIVE(2, MPI_ERR_OTHER, size)
CHECK_NEGATIVE(3, MPI_ERR_OTHER, disp_unit)
- smpi_bench_end();
+ const SmpiBenchGuard suspend_bench;
if (base == nullptr && size != 0){
retval= MPI_ERR_OTHER;
}else{
*win = new simgrid::smpi::Win( base, size, disp_unit, info, comm);
retval = MPI_SUCCESS;
}
- smpi_bench_begin();
return retval;
}
CHECK_NEGATIVE(2, MPI_ERR_OTHER, size)
CHECK_NEGATIVE(3, MPI_ERR_OTHER, disp_unit)
void* ptr = xbt_malloc(size);
- smpi_bench_end();
+ const SmpiBenchGuard suspend_bench;
*static_cast<void**>(base) = ptr;
- *win = new simgrid::smpi::Win( ptr, size, disp_unit, info, comm,1);
- smpi_bench_begin();
+ *win = new simgrid::smpi::Win(ptr, size, disp_unit, info, comm, true);
return MPI_SUCCESS;
}
if(rank==0){
ptr = xbt_malloc(size*comm->size());
}
- smpi_bench_end();
+ const SmpiBenchGuard suspend_bench;
simgrid::smpi::colls::bcast(&ptr, sizeof(void*), MPI_BYTE, 0, comm);
simgrid::smpi::colls::barrier(comm);
*static_cast<void**>(base) = (char*)ptr+rank*size;
- *win = new simgrid::smpi::Win( ptr, size, disp_unit, info, comm,rank==0);
- smpi_bench_begin();
+ *win = new simgrid::smpi::Win(ptr, size, disp_unit, info, comm, rank == 0);
return MPI_SUCCESS;
}
int PMPI_Win_create_dynamic( MPI_Info info, MPI_Comm comm, MPI_Win *win){
CHECK_COMM(2)
- smpi_bench_end();
+ const SmpiBenchGuard suspend_bench;
*win = new simgrid::smpi::Win(info, comm);
- smpi_bench_begin();
return MPI_SUCCESS;
}
CHECK_NEGATIVE(3, MPI_ERR_OTHER, size)
if (base == nullptr && size != 0)
return MPI_ERR_OTHER;
- smpi_bench_end();
+ const SmpiBenchGuard suspend_bench;
int retval = win->attach(base, size);
- smpi_bench_begin();
return retval;
}
{
CHECK_WIN(1, win)
CHECK_NULL(2, MPI_ERR_OTHER, base)
- smpi_bench_end();
+ const SmpiBenchGuard suspend_bench;
int retval = win->detach(base);
- smpi_bench_begin();
return retval;
}
int PMPI_Win_free( MPI_Win* win){
CHECK_NULL(1, MPI_ERR_WIN, win)
CHECK_WIN(1, (*win))
- smpi_bench_end();
+ if (_smpi_cfg_pedantic && (*win)->opened() == 1){//only check in pedantic mode, as it's not clear this is illegal
+ XBT_WARN("Attempt to destroy a MPI_Win too early -missing MPI_Win_fence ?");
+ return MPI_ERR_WIN;
+ }
+ const SmpiBenchGuard suspend_bench;
delete *win;
- smpi_bench_begin();
return MPI_SUCCESS;
}
{
CHECK_WIN(1, win)
CHECK_NULL(2, MPI_ERR_ARG, info)
- *info = win->info();
+ *info = new simgrid::smpi::Info(win->info());
return MPI_SUCCESS;
}
int PMPI_Win_get_group(MPI_Win win, MPI_Group * group){
CHECK_WIN(1, win)
win->get_group(group);
- (*group)->ref();
+ if (*group != MPI_COMM_WORLD->group() && *group != MPI_GROUP_NULL && *group != MPI_GROUP_EMPTY)
+ (*group)->ref();
return MPI_SUCCESS;
}
int PMPI_Win_fence( int assert, MPI_Win win){
CHECK_WIN(2, win)
- smpi_bench_end();
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
+ const SmpiBenchGuard suspend_bench;
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
TRACE_smpi_comm_in(my_proc_id, __func__, new simgrid::instr::NoOpTIData("Win_fence"));
int retval = win->fence(assert);
TRACE_smpi_comm_out(my_proc_id);
- smpi_bench_begin();
return retval;
}
int PMPI_Get( void *origin_addr, int origin_count, MPI_Datatype origin_datatype, int target_rank,
MPI_Aint target_disp, int target_count, MPI_Datatype target_datatype, MPI_Win win){
- CHECK_RMA
CHECK_WIN(8, win)
+ CHECK_RMA
+ CHECK_TARGET_DISP(5)
int retval = 0;
- smpi_bench_end();
- if (win->dynamic()==0 && target_disp <0){
- //in case of dynamic window, target_disp can be mistakenly seen as negative, as it is an address
- retval = MPI_ERR_ARG;
- } else {
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
- MPI_Group group;
- win->get_group(&group);
- TRACE_smpi_comm_in(my_proc_id, __func__,
- new simgrid::instr::Pt2PtTIData("Get", target_rank, origin_datatype->is_replayable()
- ? origin_count
- : origin_count * origin_datatype->size(),
- simgrid::smpi::Datatype::encode(origin_datatype)));
-
- retval = win->get( origin_addr, origin_count, origin_datatype, target_rank, target_disp, target_count,
- target_datatype);
- TRACE_smpi_comm_out(my_proc_id);
- }
- smpi_bench_begin();
+ const SmpiBenchGuard suspend_bench;
+
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
+ MPI_Group group;
+ win->get_group(&group);
+ TRACE_smpi_comm_in(my_proc_id, __func__,
+ new simgrid::instr::Pt2PtTIData("Get", target_rank, origin_datatype->is_replayable()
+ ? origin_count
+ : origin_count * origin_datatype->size(),
+ simgrid::smpi::Datatype::encode(origin_datatype)));
+ retval = win->get( origin_addr, origin_count, origin_datatype, target_rank, target_disp, target_count,
+ target_datatype);
+ TRACE_smpi_comm_out(my_proc_id);
+
return retval;
}
MPI_Aint target_disp, int target_count, MPI_Datatype target_datatype, MPI_Win win, MPI_Request* request){
if(target_rank==MPI_PROC_NULL)
*request = MPI_REQUEST_NULL;
- CHECK_RMA
CHECK_WIN(8, win)
+ CHECK_RMA
+ CHECK_TARGET_DISP(5)
CHECK_NULL(9, MPI_ERR_ARG, request)
int retval = 0;
- smpi_bench_end();
- if (win->dynamic()==0 && target_disp <0){
- //in case of dynamic window, target_disp can be mistakenly seen as negative, as it is an address
- retval = MPI_ERR_ARG;
- } else {
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
- MPI_Group group;
- win->get_group(&group);
- TRACE_smpi_comm_in(my_proc_id, __func__,
- new simgrid::instr::Pt2PtTIData(
- "Rget", target_rank,
- origin_datatype->is_replayable() ? origin_count : origin_count * origin_datatype->size(),
- simgrid::smpi::Datatype::encode(origin_datatype)));
-
- retval = win->get( origin_addr, origin_count, origin_datatype, target_rank, target_disp, target_count,
- target_datatype, request);
+ const SmpiBenchGuard suspend_bench;
+
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
+ MPI_Group group;
+ win->get_group(&group);
+ TRACE_smpi_comm_in(my_proc_id, __func__,
+ new simgrid::instr::Pt2PtTIData(
+ "Rget", target_rank,
+ origin_datatype->is_replayable() ? origin_count : origin_count * origin_datatype->size(),
+ simgrid::smpi::Datatype::encode(origin_datatype)));
+
+ retval = win->get( origin_addr, origin_count, origin_datatype, target_rank, target_disp, target_count,
+ target_datatype, request);
+
+ TRACE_smpi_comm_out(my_proc_id);
- TRACE_smpi_comm_out(my_proc_id);
- }
- smpi_bench_begin();
return retval;
}
int PMPI_Put(const void *origin_addr, int origin_count, MPI_Datatype origin_datatype, int target_rank,
MPI_Aint target_disp, int target_count, MPI_Datatype target_datatype, MPI_Win win){
- CHECK_RMA
CHECK_WIN(8, win)
+ CHECK_RMA
+ CHECK_TARGET_DISP(5)
int retval = 0;
- smpi_bench_end();
- if (win->dynamic()==0 && target_disp <0){
- //in case of dynamic window, target_disp can be mistakenly seen as negative, as it is an address
- retval = MPI_ERR_ARG;
- } else {
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
- MPI_Group group;
- win->get_group(&group);
- int dst_traced = group->actor(target_rank)->get_pid();
- TRACE_smpi_comm_in(my_proc_id, __func__,
- new simgrid::instr::Pt2PtTIData("Put", target_rank, origin_datatype->is_replayable()
- ? origin_count
- : origin_count * origin_datatype->size(),
- simgrid::smpi::Datatype::encode(origin_datatype)));
- TRACE_smpi_send(my_proc_id, my_proc_id, dst_traced, SMPI_RMA_TAG, origin_count * origin_datatype->size());
-
- retval = win->put( origin_addr, origin_count, origin_datatype, target_rank, target_disp, target_count,
- target_datatype);
+ const SmpiBenchGuard suspend_bench;
+
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
+ MPI_Group group;
+ win->get_group(&group);
+ aid_t dst_traced = group->actor(target_rank);
+ TRACE_smpi_comm_in(my_proc_id, __func__,
+ new simgrid::instr::Pt2PtTIData("Put", target_rank, origin_datatype->is_replayable()
+ ? origin_count
+ : origin_count * origin_datatype->size(),
+ simgrid::smpi::Datatype::encode(origin_datatype)));
+ TRACE_smpi_send(my_proc_id, my_proc_id, dst_traced, SMPI_RMA_TAG, origin_count * origin_datatype->size());
+
+ retval = win->put( origin_addr, origin_count, origin_datatype, target_rank, target_disp, target_count,
+ target_datatype);
+
+ TRACE_smpi_comm_out(my_proc_id);
- TRACE_smpi_comm_out(my_proc_id);
- }
- smpi_bench_begin();
return retval;
}
MPI_Aint target_disp, int target_count, MPI_Datatype target_datatype, MPI_Win win, MPI_Request* request){
if(target_rank==MPI_PROC_NULL)
*request = MPI_REQUEST_NULL;
- CHECK_RMA
CHECK_WIN(8, win)
+ CHECK_RMA
+ CHECK_TARGET_DISP(5)
CHECK_NULL(9, MPI_ERR_ARG, request)
int retval = 0;
- smpi_bench_end();
- if (win->dynamic()==0 && target_disp <0){
- //in case of dynamic window, target_disp can be mistakenly seen as negative, as it is an address
- retval = MPI_ERR_ARG;
- } else {
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
- MPI_Group group;
- win->get_group(&group);
- int dst_traced = group->actor(target_rank)->get_pid();
- TRACE_smpi_comm_in(my_proc_id, __func__,
- new simgrid::instr::Pt2PtTIData(
- "Rput", target_rank,
- origin_datatype->is_replayable() ? origin_count : origin_count * origin_datatype->size(),
- simgrid::smpi::Datatype::encode(origin_datatype)));
- TRACE_smpi_send(my_proc_id, my_proc_id, dst_traced, SMPI_RMA_TAG, origin_count * origin_datatype->size());
-
- retval = win->put( origin_addr, origin_count, origin_datatype, target_rank, target_disp, target_count,
- target_datatype, request);
+ const SmpiBenchGuard suspend_bench;
+
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
+ MPI_Group group;
+ win->get_group(&group);
+ aid_t dst_traced = group->actor(target_rank);
+ TRACE_smpi_comm_in(my_proc_id, __func__,
+ new simgrid::instr::Pt2PtTIData(
+ "Rput", target_rank,
+ origin_datatype->is_replayable() ? origin_count : origin_count * origin_datatype->size(),
+ simgrid::smpi::Datatype::encode(origin_datatype)));
+ TRACE_smpi_send(my_proc_id, my_proc_id, dst_traced, SMPI_RMA_TAG, origin_count * origin_datatype->size());
+
+ retval = win->put( origin_addr, origin_count, origin_datatype, target_rank, target_disp, target_count,
+ target_datatype, request);
+
+ TRACE_smpi_comm_out(my_proc_id);
- TRACE_smpi_comm_out(my_proc_id);
- }
- smpi_bench_begin();
return retval;
}
int PMPI_Accumulate(const void *origin_addr, int origin_count, MPI_Datatype origin_datatype, int target_rank,
MPI_Aint target_disp, int target_count, MPI_Datatype target_datatype, MPI_Op op, MPI_Win win){
- CHECK_RMA
- CHECK_OP(8)
CHECK_WIN(9, win)
+ CHECK_RMA
+ CHECK_MPI_NULL(8, MPI_OP_NULL, MPI_ERR_OP, op)
+ CHECK_TARGET_DISP(5)
int retval = 0;
- smpi_bench_end();
- if (win->dynamic()==0 && target_disp <0){
- //in case of dynamic window, target_disp can be mistakenly seen as negative, as it is an address
- retval = MPI_ERR_ARG;
- } else {
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
- MPI_Group group;
- win->get_group(&group);
- TRACE_smpi_comm_in(my_proc_id, __func__,
- new simgrid::instr::Pt2PtTIData(
- "Accumulate", target_rank,
- origin_datatype->is_replayable() ? origin_count : origin_count * origin_datatype->size(),
- simgrid::smpi::Datatype::encode(origin_datatype)));
- retval = win->accumulate( origin_addr, origin_count, origin_datatype, target_rank, target_disp, target_count,
- target_datatype, op);
- TRACE_smpi_comm_out(my_proc_id);
- }
- smpi_bench_begin();
+ const SmpiBenchGuard suspend_bench;
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
+ MPI_Group group;
+ win->get_group(&group);
+ TRACE_smpi_comm_in(my_proc_id, __func__,
+ new simgrid::instr::Pt2PtTIData(
+ "Accumulate", target_rank,
+ origin_datatype->is_replayable() ? origin_count : origin_count * origin_datatype->size(),
+ simgrid::smpi::Datatype::encode(origin_datatype)));
+ retval = win->accumulate( origin_addr, origin_count, origin_datatype, target_rank, target_disp, target_count,
+ target_datatype, op);
+
+ TRACE_smpi_comm_out(my_proc_id);
+
return retval;
}
MPI_Aint target_disp, int target_count, MPI_Datatype target_datatype, MPI_Op op, MPI_Win win, MPI_Request* request){
if(target_rank==MPI_PROC_NULL)
*request = MPI_REQUEST_NULL;
- CHECK_RMA
- CHECK_OP(8)
CHECK_WIN(9, win)
+ CHECK_RMA
+ CHECK_MPI_NULL(8, MPI_OP_NULL, MPI_ERR_OP, op)
+ CHECK_TARGET_DISP(5)
CHECK_NULL(10, MPI_ERR_ARG, request)
int retval = 0;
- smpi_bench_end();
- if (win->dynamic()==0 && target_disp <0){
- //in case of dynamic window, target_disp can be mistakenly seen as negative, as it is an address
- retval = MPI_ERR_ARG;
- } else {
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
- MPI_Group group;
- win->get_group(&group);
- TRACE_smpi_comm_in(my_proc_id, __func__,
- new simgrid::instr::Pt2PtTIData(
- "Raccumulate", target_rank,
- origin_datatype->is_replayable() ? origin_count : origin_count * origin_datatype->size(),
- simgrid::smpi::Datatype::encode(origin_datatype)));
-
- retval = win->accumulate( origin_addr, origin_count, origin_datatype, target_rank, target_disp, target_count,
- target_datatype, op, request);
- TRACE_smpi_comm_out(my_proc_id);
- }
- smpi_bench_begin();
+ const SmpiBenchGuard suspend_bench;
+
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
+ MPI_Group group;
+ win->get_group(&group);
+ TRACE_smpi_comm_in(my_proc_id, __func__,
+ new simgrid::instr::Pt2PtTIData(
+ "Raccumulate", target_rank,
+ origin_datatype->is_replayable() ? origin_count : origin_count * origin_datatype->size(),
+ simgrid::smpi::Datatype::encode(origin_datatype)));
+
+ retval = win->accumulate( origin_addr, origin_count, origin_datatype, target_rank, target_disp, target_count,
+ target_datatype, op, request);
+
+ TRACE_smpi_comm_out(my_proc_id);
+
return retval;
}
-int PMPI_Get_accumulate(const void *origin_addr, int origin_count, MPI_Datatype origin_datatype, void *result_addr,
-int result_count, MPI_Datatype result_datatype, int target_rank, MPI_Aint target_disp, int target_count,
-MPI_Datatype target_datatype, MPI_Op op, MPI_Win win){
- if (op != MPI_NO_OP)
- CHECK_BUFFER(1, origin_addr, origin_count)
- CHECK_COUNT(2, origin_count)
- if(origin_count>0)
- CHECK_TYPE(3, origin_datatype)
- CHECK_BUFFER(4, result_addr, result_count)
+int PMPI_Get_accumulate(const void* origin_addr, int origin_count, MPI_Datatype origin_datatype, void* result_addr,
+ int result_count, MPI_Datatype result_datatype, int target_rank, MPI_Aint target_disp,
+ int target_count, MPI_Datatype target_datatype, MPI_Op op, MPI_Win win)
+{
+ if (op != MPI_NO_OP){
+ CHECK_BUFFER(1, origin_addr, origin_count, origin_datatype)
+ CHECK_COUNT(2, origin_count)
+ if(origin_count>0)
+ CHECK_TYPE(3, origin_datatype)
+ }
CHECK_COUNT(5, result_count)
CHECK_TYPE(6, result_datatype)
- CHECK_PROC(7, target_rank)
- CHECK_NEGATIVE(7, MPI_ERR_RANK, target_rank)
+ CHECK_BUFFER(4, result_addr, result_count, result_datatype)
+ CHECK_WIN(12, win)
+ CHECK_PROC_RMA(7, target_rank, win)
CHECK_COUNT(9, target_count)
CHECK_TYPE(10, target_datatype)
- CHECK_OP(11)
- CHECK_WIN(12, win)
+ CHECK_MPI_NULL(11, MPI_OP_NULL, MPI_ERR_OP, op)
+ CHECK_TARGET_DISP(8)
+
int retval = 0;
- smpi_bench_end();
- if (win->dynamic()==0 && target_disp <0){
- //in case of dynamic window, target_disp can be mistakenly seen as negative, as it is an address
- retval = MPI_ERR_ARG;
- } else {
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
- MPI_Group group;
- win->get_group(&group);
- TRACE_smpi_comm_in(my_proc_id, __func__,
- new simgrid::instr::Pt2PtTIData(
- "Get_accumulate", target_rank,
- target_datatype->is_replayable() ? target_count : target_count * target_datatype->size(),
- simgrid::smpi::Datatype::encode(target_datatype)));
-
- retval = win->get_accumulate( origin_addr, origin_count, origin_datatype, result_addr,
- result_count, result_datatype, target_rank, target_disp,
- target_count, target_datatype, op);
+ const SmpiBenchGuard suspend_bench;
+
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
+ MPI_Group group;
+ win->get_group(&group);
+ TRACE_smpi_comm_in(my_proc_id, __func__,
+ new simgrid::instr::Pt2PtTIData(
+ "Get_accumulate", target_rank,
+ target_datatype->is_replayable() ? target_count : target_count * target_datatype->size(),
+ simgrid::smpi::Datatype::encode(target_datatype)));
+
+ retval = win->get_accumulate( origin_addr, origin_count, origin_datatype, result_addr,
+ result_count, result_datatype, target_rank, target_disp,
+ target_count, target_datatype, op);
+
+ TRACE_smpi_comm_out(my_proc_id);
- TRACE_smpi_comm_out(my_proc_id);
- }
- smpi_bench_begin();
return retval;
}
-
-int PMPI_Rget_accumulate(const void *origin_addr, int origin_count, MPI_Datatype origin_datatype, void *result_addr,
-int result_count, MPI_Datatype result_datatype, int target_rank, MPI_Aint target_disp, int target_count,
-MPI_Datatype target_datatype, MPI_Op op, MPI_Win win, MPI_Request* request){
+int PMPI_Rget_accumulate(const void* origin_addr, int origin_count, MPI_Datatype origin_datatype, void* result_addr,
+ int result_count, MPI_Datatype result_datatype, int target_rank, MPI_Aint target_disp,
+ int target_count, MPI_Datatype target_datatype, MPI_Op op, MPI_Win win, MPI_Request* request)
+{
if(target_rank==MPI_PROC_NULL)
*request = MPI_REQUEST_NULL;
- CHECK_BUFFER(1, origin_addr, origin_count)
- CHECK_COUNT(2, origin_count)
- CHECK_TYPE(3, origin_datatype)
- CHECK_BUFFER(4, result_addr, result_count)
+ if (op != MPI_NO_OP){
+ CHECK_BUFFER(1, origin_addr, origin_count, origin_datatype)
+ CHECK_COUNT(2, origin_count)
+ if(origin_count>0)
+ CHECK_TYPE(3, origin_datatype)
+ }
CHECK_COUNT(5, result_count)
CHECK_TYPE(6, result_datatype)
- CHECK_PROC(7, target_rank)
- CHECK_NEGATIVE(7, MPI_ERR_RANK, target_rank)
+ CHECK_BUFFER(4, result_addr, result_count, result_datatype)
+ CHECK_WIN(12, win)
+ CHECK_PROC_RMA(7, target_rank, win)
CHECK_COUNT(9, target_count)
CHECK_TYPE(10, target_datatype)
- CHECK_OP(11)
- CHECK_WIN(12, win)
+ CHECK_MPI_NULL(11, MPI_OP_NULL, MPI_ERR_OP, op)
+ CHECK_TARGET_DISP(8)
CHECK_NULL(10, MPI_ERR_ARG, request)
int retval = 0;
- smpi_bench_end();
- if (win->dynamic()==0 && target_disp <0){
- //in case of dynamic window, target_disp can be mistakenly seen as negative, as it is an address
- retval = MPI_ERR_ARG;
- } else {
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
- MPI_Group group;
- win->get_group(&group);
- TRACE_smpi_comm_in(my_proc_id, __func__,
- new simgrid::instr::Pt2PtTIData(
- "Rget_accumulate", target_rank,
- target_datatype->is_replayable() ? target_count : target_count * target_datatype->size(),
- simgrid::smpi::Datatype::encode(target_datatype)));
-
- retval = win->get_accumulate( origin_addr, origin_count, origin_datatype, result_addr,
- result_count, result_datatype, target_rank, target_disp,
- target_count, target_datatype, op, request);
+ const SmpiBenchGuard suspend_bench;
+
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
+ MPI_Group group;
+ win->get_group(&group);
+ TRACE_smpi_comm_in(my_proc_id, __func__,
+ new simgrid::instr::Pt2PtTIData(
+ "Rget_accumulate", target_rank,
+ target_datatype->is_replayable() ? target_count : target_count * target_datatype->size(),
+ simgrid::smpi::Datatype::encode(target_datatype)));
+
+ retval = win->get_accumulate( origin_addr, origin_count, origin_datatype, result_addr,
+ result_count, result_datatype, target_rank, target_disp,
+ target_count, target_datatype, op, request);
+
+ TRACE_smpi_comm_out(my_proc_id);
- TRACE_smpi_comm_out(my_proc_id);
- }
- smpi_bench_begin();
return retval;
}
CHECK_NULL(2, MPI_ERR_BUFFER, compare_addr)
CHECK_NULL(3, MPI_ERR_BUFFER, result_addr)
CHECK_TYPE(4, datatype)
- CHECK_PROC(5, target_rank)
- CHECK_NEGATIVE(5, MPI_ERR_RANK, target_rank)
CHECK_WIN(6, win)
+ CHECK_PROC_RMA(5, target_rank, win)
+ CHECK_TARGET_DISP(6)
+
int retval = 0;
- smpi_bench_end();
- if (win->dynamic()==0 && target_disp <0){
- //in case of dynamic window, target_disp can be mistakenly seen as negative, as it is an address
- retval = MPI_ERR_ARG;
- } else {
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
- MPI_Group group;
- win->get_group(&group);
- TRACE_smpi_comm_in(my_proc_id, __func__,
- new simgrid::instr::Pt2PtTIData("Compare_and_swap", target_rank,
- datatype->is_replayable() ? 1 : datatype->size(),
- simgrid::smpi::Datatype::encode(datatype)));
- retval = win->compare_and_swap(origin_addr, compare_addr, result_addr, datatype, target_rank, target_disp);
+ const SmpiBenchGuard suspend_bench;
+
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
+ MPI_Group group;
+ win->get_group(&group);
+ TRACE_smpi_comm_in(my_proc_id, __func__,
+ new simgrid::instr::Pt2PtTIData("Compare_and_swap", target_rank,
+ datatype->is_replayable() ? 1 : datatype->size(),
+ simgrid::smpi::Datatype::encode(datatype)));
+
+ retval = win->compare_and_swap(origin_addr, compare_addr, result_addr, datatype, target_rank, target_disp);
+
+ TRACE_smpi_comm_out(my_proc_id);
- TRACE_smpi_comm_out(my_proc_id);
- }
- smpi_bench_begin();
return retval;
}
int PMPI_Win_post(MPI_Group group, int assert, MPI_Win win){
CHECK_GROUP(1, group)
CHECK_WIN(2, win)
- smpi_bench_end();
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
+ const SmpiBenchGuard suspend_bench;
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
TRACE_smpi_comm_in(my_proc_id, __func__, new simgrid::instr::NoOpTIData("Win_post"));
int retval = win->post(group,assert);
TRACE_smpi_comm_out(my_proc_id);
- smpi_bench_begin();
return retval;
}
int PMPI_Win_start(MPI_Group group, int assert, MPI_Win win){
CHECK_GROUP(1, group)
CHECK_WIN(2, win)
- smpi_bench_end();
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
+ const SmpiBenchGuard suspend_bench;
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
TRACE_smpi_comm_in(my_proc_id, __func__, new simgrid::instr::NoOpTIData("Win_start"));
int retval = win->start(group,assert);
TRACE_smpi_comm_out(my_proc_id);
- smpi_bench_begin();
return retval;
}
int PMPI_Win_complete(MPI_Win win){
CHECK_WIN(1, win)
- smpi_bench_end();
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
+ const SmpiBenchGuard suspend_bench;
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
TRACE_smpi_comm_in(my_proc_id, __func__, new simgrid::instr::NoOpTIData("Win_complete"));
int retval = win->complete();
TRACE_smpi_comm_out(my_proc_id);
- smpi_bench_begin();
return retval;
}
int PMPI_Win_wait(MPI_Win win){
CHECK_WIN(1, win)
- smpi_bench_end();
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
+ const SmpiBenchGuard suspend_bench;
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
TRACE_smpi_comm_in(my_proc_id, __func__, new simgrid::instr::NoOpTIData("Win_wait"));
int retval = win->wait();
TRACE_smpi_comm_out(my_proc_id);
- smpi_bench_begin();
return retval;
}
int PMPI_Win_lock(int lock_type, int rank, int assert, MPI_Win win){
- CHECK_PROC(2, rank)
CHECK_WIN(4, win)
+ CHECK_PROC_RMA(2, rank, win)
int retval = 0;
- smpi_bench_end();
+ const SmpiBenchGuard suspend_bench;
if (lock_type != MPI_LOCK_EXCLUSIVE &&
lock_type != MPI_LOCK_SHARED) {
retval = MPI_ERR_LOCKTYPE;
} else {
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
TRACE_smpi_comm_in(my_proc_id, __func__, new simgrid::instr::NoOpTIData("Win_lock"));
retval = win->lock(lock_type,rank,assert);
TRACE_smpi_comm_out(my_proc_id);
}
- smpi_bench_begin();
return retval;
}
int PMPI_Win_unlock(int rank, MPI_Win win){
- CHECK_PROC(1, rank)
CHECK_WIN(2, win)
- smpi_bench_end();
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
+ CHECK_PROC_RMA(1, rank, win)
+ const SmpiBenchGuard suspend_bench;
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
TRACE_smpi_comm_in(my_proc_id, __func__, new simgrid::instr::NoOpTIData("Win_unlock"));
int retval = win->unlock(rank);
TRACE_smpi_comm_out(my_proc_id);
- smpi_bench_begin();
return retval;
}
int PMPI_Win_lock_all(int assert, MPI_Win win){
CHECK_WIN(2, win)
- smpi_bench_end();
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
+ const SmpiBenchGuard suspend_bench;
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
TRACE_smpi_comm_in(my_proc_id, __func__, new simgrid::instr::NoOpTIData("Win_lock_all"));
int retval = win->lock_all(assert);
TRACE_smpi_comm_out(my_proc_id);
- smpi_bench_begin();
return retval;
}
int PMPI_Win_unlock_all(MPI_Win win){
CHECK_WIN(1, win)
- smpi_bench_end();
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
+ const SmpiBenchGuard suspend_bench;
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
TRACE_smpi_comm_in(my_proc_id, __func__, new simgrid::instr::NoOpTIData("Win_unlock_all"));
int retval = win->unlock_all();
TRACE_smpi_comm_out(my_proc_id);
- smpi_bench_begin();
return retval;
}
int PMPI_Win_flush(int rank, MPI_Win win){
- CHECK_PROC(1, rank)
CHECK_WIN(2, win)
- smpi_bench_end();
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
+ CHECK_PROC_RMA(1, rank, win)
+ const SmpiBenchGuard suspend_bench;
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
TRACE_smpi_comm_in(my_proc_id, __func__, new simgrid::instr::NoOpTIData("Win_flush"));
int retval = win->flush(rank);
TRACE_smpi_comm_out(my_proc_id);
- smpi_bench_begin();
return retval;
}
int PMPI_Win_flush_local(int rank, MPI_Win win){
- CHECK_PROC(1, rank)
- CHECK_WIN(2, win) smpi_bench_end();
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
+ CHECK_WIN(2, win)
+ CHECK_PROC_RMA(1, rank, win)
+ const SmpiBenchGuard suspend_bench;
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
TRACE_smpi_comm_in(my_proc_id, __func__, new simgrid::instr::NoOpTIData("Win_flush_local"));
int retval = win->flush_local(rank);
TRACE_smpi_comm_out(my_proc_id);
- smpi_bench_begin();
return retval;
}
int PMPI_Win_flush_all(MPI_Win win){
CHECK_WIN(1, win)
- smpi_bench_end();
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
+ const SmpiBenchGuard suspend_bench;
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
TRACE_smpi_comm_in(my_proc_id, __func__, new simgrid::instr::NoOpTIData("Win_flush_all"));
int retval = win->flush_all();
TRACE_smpi_comm_out(my_proc_id);
- smpi_bench_begin();
return retval;
}
int PMPI_Win_flush_local_all(MPI_Win win){
CHECK_WIN(1, win)
- smpi_bench_end();
- int my_proc_id = simgrid::s4u::this_actor::get_pid();
+ const SmpiBenchGuard suspend_bench;
+ aid_t my_proc_id = simgrid::s4u::this_actor::get_pid();
TRACE_smpi_comm_in(my_proc_id, __func__, new simgrid::instr::NoOpTIData("Win_flush_local_all"));
int retval = win->flush_local_all();
TRACE_smpi_comm_out(my_proc_id);
- smpi_bench_begin();
return retval;
}
