1 /* Copyright (c) 2007-2015. The SimGrid Team.
2 * All rights reserved. */
4 /* This program is free software; you can redistribute it and/or modify it
5 * under the terms of the license (GNU LGPL) which comes with this package. */
10 XBT_LOG_NEW_DEFAULT_SUBCATEGORY(smpi_rma, smpi, "Logging specific to SMPI (RMA operations)");
14 std::unordered_map<int, smpi_key_elem> Win::keyvals_;
15 int Win::keyval_id_=0;
17 Win::Win(void *base, MPI_Aint size, int disp_unit, MPI_Info info, MPI_Comm comm): base_(base), size_(size), disp_unit_(disp_unit), assert_(0), info_(info), comm_(comm){
18 int comm_size = comm->size();
20 XBT_DEBUG("Creating window");
21 if(info!=MPI_INFO_NULL)
25 group_ = MPI_GROUP_NULL;
26 requests_ = new std::vector<MPI_Request>();
27 mut_=xbt_mutex_init();
28 lock_mut_=xbt_mutex_init();
29 connected_wins_ = new MPI_Win[comm_size];
30 connected_wins_[rank_] = this;
33 bar_ = MSG_barrier_init(comm_size);
37 comm->add_rma_win(this);
39 Colls::allgather(&(connected_wins_[rank_]), sizeof(MPI_Win), MPI_BYTE, connected_wins_, sizeof(MPI_Win),
42 Colls::bcast(&(bar_), sizeof(msg_bar_t), MPI_BYTE, 0, comm);
48 //As per the standard, perform a barrier to ensure every async comm is finished
49 MSG_barrier_wait(bar_);
51 int finished = finish_comms();
52 XBT_DEBUG("Win destructor - Finished %d RMA calls", finished);
55 delete[] connected_wins_;
56 if (name_ != nullptr){
59 if(info_!=MPI_INFO_NULL){
60 MPI_Info_free(&info_);
63 comm_->remove_rma_win(this);
65 Colls::barrier(comm_);
66 int rank=comm_->rank();
68 MSG_barrier_destroy(bar_);
69 xbt_mutex_destroy(mut_);
70 xbt_mutex_destroy(lock_mut_);
75 void Win::get_name(char* name, int* length){
81 *length = strlen(name_);
82 strncpy(name, name_, *length+1);
85 void Win::get_group(MPI_Group* group){
86 if(comm_ != MPI_COMM_NULL){
87 *group = comm_->group();
89 *group = MPI_GROUP_NULL;
105 int Win::disp_unit(){
110 void Win::set_name(char* name){
111 name_ = xbt_strdup(name);
114 int Win::fence(int assert)
116 XBT_DEBUG("Entering fence");
119 if (assert != MPI_MODE_NOPRECEDE) {
120 // This is not the first fence => finalize what came before
121 MSG_barrier_wait(bar_);
122 xbt_mutex_acquire(mut_);
123 // This (simulated) mutex ensures that no process pushes to the vector of requests during the waitall.
124 // Without this, the vector could get redimensionned when another process pushes.
125 // This would result in the array used by Request::waitall() to be invalidated.
126 // Another solution would be to copy the data and cleanup the vector *before* Request::waitall
127 std::vector<MPI_Request> *reqs = requests_;
128 int size = static_cast<int>(reqs->size());
129 // start all requests that have been prepared by another process
131 for (const auto& req : *reqs) {
132 if (req && (req->flags() & PREPARED))
136 MPI_Request* treqs = &(*reqs)[0];
138 Request::waitall(size, treqs, MPI_STATUSES_IGNORE);
141 xbt_mutex_release(mut_);
145 MSG_barrier_wait(bar_);
146 XBT_DEBUG("Leaving fence");
151 int Win::put( void *origin_addr, int origin_count, MPI_Datatype origin_datatype, int target_rank,
152 MPI_Aint target_disp, int target_count, MPI_Datatype target_datatype)
154 //get receiver pointer
155 MPI_Win recv_win = connected_wins_[target_rank];
157 if(opened_==0){//check that post/start has been done
158 // no fence or start .. lock ok ?
160 for(auto it : recv_win->lockers_)
161 if (it == comm_->rank())
167 if(target_count*target_datatype->get_extent()>recv_win->size_)
170 void* recv_addr = static_cast<void*> ( static_cast<char*>(recv_win->base_) + target_disp * recv_win->disp_unit_);
171 XBT_DEBUG("Entering MPI_Put to %d", target_rank);
173 if(target_rank != comm_->rank()){
174 //prepare send_request
175 MPI_Request sreq = Request::rma_send_init(origin_addr, origin_count, origin_datatype, smpi_process()->index(),
176 comm_->group()->index(target_rank), SMPI_RMA_TAG+1, comm_, MPI_OP_NULL);
178 //prepare receiver request
179 MPI_Request rreq = Request::rma_recv_init(recv_addr, target_count, target_datatype, smpi_process()->index(),
180 comm_->group()->index(target_rank), SMPI_RMA_TAG+1, recv_win->comm_, MPI_OP_NULL);
182 //push request to receiver's win
183 xbt_mutex_acquire(recv_win->mut_);
184 recv_win->requests_->push_back(rreq);
185 xbt_mutex_release(recv_win->mut_);
189 //push request to sender's win
190 xbt_mutex_acquire(mut_);
191 requests_->push_back(sreq);
192 xbt_mutex_release(mut_);
194 Datatype::copy(origin_addr, origin_count, origin_datatype, recv_addr, target_count, target_datatype);
200 int Win::get( void *origin_addr, int origin_count, MPI_Datatype origin_datatype, int target_rank,
201 MPI_Aint target_disp, int target_count, MPI_Datatype target_datatype)
204 MPI_Win send_win = connected_wins_[target_rank];
206 if(opened_==0){//check that post/start has been done
207 // no fence or start .. lock ok ?
209 for(auto it : send_win->lockers_)
210 if (it == comm_->rank())
216 if(target_count*target_datatype->get_extent()>send_win->size_)
219 void* send_addr = static_cast<void*>(static_cast<char*>(send_win->base_) + target_disp * send_win->disp_unit_);
220 XBT_DEBUG("Entering MPI_Get from %d", target_rank);
222 if(target_rank != comm_->rank()){
223 //prepare send_request
224 MPI_Request sreq = Request::rma_send_init(send_addr, target_count, target_datatype,
225 comm_->group()->index(target_rank), smpi_process()->index(), SMPI_RMA_TAG+2, send_win->comm_,
228 //prepare receiver request
229 MPI_Request rreq = Request::rma_recv_init(origin_addr, origin_count, origin_datatype,
230 comm_->group()->index(target_rank), smpi_process()->index(), SMPI_RMA_TAG+2, comm_,
233 //start the send, with another process than us as sender.
235 //push request to receiver's win
236 xbt_mutex_acquire(send_win->mut_);
237 send_win->requests_->push_back(sreq);
238 xbt_mutex_release(send_win->mut_);
242 //push request to sender's win
243 xbt_mutex_acquire(mut_);
244 requests_->push_back(rreq);
245 xbt_mutex_release(mut_);
247 Datatype::copy(send_addr, target_count, target_datatype, origin_addr, origin_count, origin_datatype);
254 int Win::accumulate( void *origin_addr, int origin_count, MPI_Datatype origin_datatype, int target_rank,
255 MPI_Aint target_disp, int target_count, MPI_Datatype target_datatype, MPI_Op op)
258 //get receiver pointer
259 MPI_Win recv_win = connected_wins_[target_rank];
261 if(opened_==0){//check that post/start has been done
262 // no fence or start .. lock ok ?
264 for(auto it : recv_win->lockers_)
265 if (it == comm_->rank())
270 //FIXME: local version
272 if(target_count*target_datatype->get_extent()>recv_win->size_)
275 void* recv_addr = static_cast<void*>(static_cast<char*>(recv_win->base_) + target_disp * recv_win->disp_unit_);
276 XBT_DEBUG("Entering MPI_Accumulate to %d", target_rank);
277 //As the tag will be used for ordering of the operations, substract count from it (to avoid collisions with other SMPI tags, SMPI_RMA_TAG is set below all the other ones we use )
278 //prepare send_request
280 MPI_Request sreq = Request::rma_send_init(origin_addr, origin_count, origin_datatype,
281 smpi_process()->index(), comm_->group()->index(target_rank), SMPI_RMA_TAG-3-count_, comm_, op);
283 //prepare receiver request
284 MPI_Request rreq = Request::rma_recv_init(recv_addr, target_count, target_datatype,
285 smpi_process()->index(), comm_->group()->index(target_rank), SMPI_RMA_TAG-3-count_, recv_win->comm_, op);
288 //push request to receiver's win
289 xbt_mutex_acquire(recv_win->mut_);
290 recv_win->requests_->push_back(rreq);
291 xbt_mutex_release(recv_win->mut_);
295 //push request to sender's win
296 xbt_mutex_acquire(mut_);
297 requests_->push_back(sreq);
298 xbt_mutex_release(mut_);
303 int Win::start(MPI_Group group, int assert){
304 /* From MPI forum advices
305 The call to MPI_WIN_COMPLETE does not return until the put call has completed at the origin; and the target window
306 will be accessed by the put operation only after the call to MPI_WIN_START has matched a call to MPI_WIN_POST by
307 the target process. This still leaves much choice to implementors. The call to MPI_WIN_START can block until the
308 matching call to MPI_WIN_POST occurs at all target processes. One can also have implementations where the call to
309 MPI_WIN_START is nonblocking, but the call to MPI_PUT blocks until the matching call to MPI_WIN_POST occurred; or
310 implementations where the first two calls are nonblocking, but the call to MPI_WIN_COMPLETE blocks until the call
311 to MPI_WIN_POST occurred; or even implementations where all three calls can complete before any target process
312 called MPI_WIN_POST --- the data put must be buffered, in this last case, so as to allow the put to complete at the
313 origin ahead of its completion at the target. However, once the call to MPI_WIN_POST is issued, the sequence above
314 must complete, without further dependencies. */
316 //naive, blocking implementation.
319 int size = group->size();
320 MPI_Request* reqs = xbt_new0(MPI_Request, size);
323 int src = group->index(j);
324 if (src != smpi_process()->index() && src != MPI_UNDEFINED) {
325 reqs[i] = Request::irecv_init(nullptr, 0, MPI_CHAR, src, SMPI_RMA_TAG + 4, MPI_COMM_WORLD);
331 Request::startall(size, reqs);
332 Request::waitall(size, reqs, MPI_STATUSES_IGNORE);
334 Request::unref(&reqs[i]);
337 opened_++; //we're open for business !
343 int Win::post(MPI_Group group, int assert){
344 //let's make a synchronous send here
347 int size = group->size();
348 MPI_Request* reqs = xbt_new0(MPI_Request, size);
351 int dst=group->index(j);
352 if(dst!=smpi_process()->index() && dst!=MPI_UNDEFINED){
353 reqs[i]=Request::send_init(nullptr, 0, MPI_CHAR, dst, SMPI_RMA_TAG+4, MPI_COMM_WORLD);
360 Request::startall(size, reqs);
361 Request::waitall(size, reqs, MPI_STATUSES_IGNORE);
363 Request::unref(&reqs[i]);
366 opened_++; //we're open for business !
374 xbt_die("Complete called on already opened MPI_Win");
376 XBT_DEBUG("Entering MPI_Win_Complete");
379 int size = group_->size();
380 MPI_Request* reqs = xbt_new0(MPI_Request, size);
383 int dst=group_->index(j);
384 if(dst!=smpi_process()->index() && dst!=MPI_UNDEFINED){
385 reqs[i]=Request::send_init(nullptr, 0, MPI_CHAR, dst, SMPI_RMA_TAG+5, MPI_COMM_WORLD);
391 XBT_DEBUG("Win_complete - Sending sync messages to %d processes", size);
392 Request::startall(size, reqs);
393 Request::waitall(size, reqs, MPI_STATUSES_IGNORE);
396 Request::unref(&reqs[i]);
400 int finished = finish_comms();
401 XBT_DEBUG("Win_complete - Finished %d RMA calls", finished);
403 Group::unref(group_);
404 opened_--; //we're closed for business !
409 //naive, blocking implementation.
410 XBT_DEBUG("Entering MPI_Win_Wait");
412 int size = group_->size();
413 MPI_Request* reqs = xbt_new0(MPI_Request, size);
416 int src=group_->index(j);
417 if(src!=smpi_process()->index() && src!=MPI_UNDEFINED){
418 reqs[i]=Request::irecv_init(nullptr, 0, MPI_CHAR, src,SMPI_RMA_TAG+5, MPI_COMM_WORLD);
424 XBT_DEBUG("Win_wait - Receiving sync messages from %d processes", size);
425 Request::startall(size, reqs);
426 Request::waitall(size, reqs, MPI_STATUSES_IGNORE);
428 Request::unref(&reqs[i]);
431 int finished = finish_comms();
432 XBT_DEBUG("Win_wait - Finished %d RMA calls", finished);
434 Group::unref(group_);
435 opened_--; //we're opened for business !
439 int Win::lock(int lock_type, int rank, int assert){
443 MPI_Win target_win = connected_wins_[rank];
445 if ((lock_type == MPI_LOCK_EXCLUSIVE && target_win->mode_ != MPI_LOCK_SHARED)|| target_win->mode_ == MPI_LOCK_EXCLUSIVE){
446 xbt_mutex_acquire(target_win->lock_mut_);
447 target_win->mode_+= lock_type;//add the lock_type to differentiate case when we are switching from EXCLUSIVE to SHARED (no release needed in the unlock)
448 if(lock_type == MPI_LOCK_SHARED){//the window used to be exclusive, it's now shared.
449 xbt_mutex_release(target_win->lock_mut_);
451 } else if(!(target_win->mode_==MPI_LOCK_SHARED && lock_type == MPI_LOCK_EXCLUSIVE))
452 target_win->mode_+= lock_type; // don't set to exclusive if it's already shared
454 target_win->lockers_.push_back(comm_->rank());
456 int finished = finish_comms();
457 XBT_DEBUG("Win_lock %d - Finished %d RMA calls", rank, finished);
462 int Win::unlock(int rank){
466 MPI_Win target_win = connected_wins_[rank];
467 int target_mode = target_win->mode_;
468 target_win->mode_= 0;
469 target_win->lockers_.remove(comm_->rank());
470 if (target_mode==MPI_LOCK_EXCLUSIVE){
471 xbt_mutex_release(target_win->lock_mut_);
474 int finished = finish_comms();
475 XBT_DEBUG("Win_unlock %d - Finished %d RMA calls", rank, finished);
480 Win* Win::f2c(int id){
481 return static_cast<Win*>(F2C::f2c(id));
485 int Win::finish_comms(){
486 xbt_mutex_acquire(mut_);
487 //Finish own requests
488 std::vector<MPI_Request> *reqqs = requests_;
489 int size = static_cast<int>(reqqs->size());
491 // start all requests that have been prepared by another process
492 for (const auto& req : *reqqs) {
493 if (req && (req->flags() & PREPARED))
497 MPI_Request* treqs = &(*reqqs)[0];
498 Request::waitall(size, treqs, MPI_STATUSES_IGNORE);
501 xbt_mutex_release(mut_);