XBT_LOG_NEW_DEFAULT_SUBCATEGORY(smpi_replay,smpi,"Trace Replay with SMPI");
static std::unordered_map<int, std::vector<MPI_Request>*> reqq;
+typedef std::tuple</*sender*/ int, /* reciever */ int, /* tag */int> req_key_t;
+typedef std::unordered_map<req_key_t, MPI_Request, hash_tuple::hash<std::tuple<int,int,int>>> req_storage_t;
static MPI_Datatype MPI_DEFAULT_TYPE;
namespace smpi {
namespace replay {
+
+class RequestStorage {
+private:
+ req_storage_t store;
+
+public:
+ RequestStorage() {}
+ int size()
+ {
+ return store.size();
+ }
+
+ req_storage_t& get_store()
+ {
+ return store;
+ }
+
+ void get_requests(std::vector<MPI_Request>& vec)
+ {
+ for (auto& pair : store) {
+ auto& req = pair.second;
+ auto my_proc_id = simgrid::s4u::this_actor::get_pid();
+ if (req != MPI_REQUEST_NULL && (req->src() == my_proc_id || req->dst() == my_proc_id)) {
+ vec.push_back(pair.second);
+ pair.second->print_request("MM");
+ }
+ }
+ }
+
+ MPI_Request find(int src, int dst, int tag)
+ {
+ req_storage_t::iterator it = store.find(req_key_t(src, dst, tag));
+ return (it == store.end()) ? MPI_REQUEST_NULL : it->second;
+ }
+
+ void remove(MPI_Request req)
+ {
+ if (req == MPI_REQUEST_NULL) return;
+
+ store.erase(req_key_t(req->src()-1, req->dst()-1, req->tag()));
+ }
+
+ void add(MPI_Request req)
+ {
+ if (req != MPI_REQUEST_NULL) // Can and does happen in the case of TestAction
+ store.insert({req_key_t(req->src()-1, req->dst()-1, req->tag()), req});
+ }
+
+ /* Sometimes we need to re-insert MPI_REQUEST_NULL but we still need src,dst and tag */
+ void addNullRequest(int src, int dst, int tag)
+ {
+ store.insert({req_key_t(src, dst, tag), MPI_REQUEST_NULL});
+ }
+};
+
class ActionArgParser {
public:
virtual ~ActionArgParser() = default;
template <class T> class ReplayAction {
protected:
const std::string name;
+ RequestStorage* req_storage; // Points to the right storage for this process, nullptr except for Send/Recv/Wait/Test actions.
const int my_proc_id;
T args;
public:
- explicit ReplayAction(std::string name) : name(name), my_proc_id(simgrid::s4u::this_actor::get_pid()) {}
+ explicit ReplayAction(std::string name, RequestStorage& storage) : name(name), req_storage(&storage), my_proc_id(simgrid::s4u::this_actor::get_pid()) {}
+ explicit ReplayAction(std::string name) : name(name), req_storage(nullptr), my_proc_id(simgrid::s4u::this_actor::get_pid()) {}
virtual ~ReplayAction() = default;
virtual void execute(simgrid::xbt::ReplayAction& action)
class WaitAction : public ReplayAction<WaitTestParser> {
public:
- WaitAction() : ReplayAction("Wait") {}
+ WaitAction(RequestStorage& storage) : ReplayAction("Wait", storage) {}
void kernel(simgrid::xbt::ReplayAction& action) override
{
std::string s = boost::algorithm::join(action, " ");
- xbt_assert(get_reqq_self()->size(), "action wait not preceded by any irecv or isend: %s", s.c_str());
- MPI_Request request = get_reqq_self()->back();
- get_reqq_self()->pop_back();
+ xbt_assert(req_storage->size(), "action wait not preceded by any irecv or isend: %s", s.c_str());
+ MPI_Request request = req_storage->find(args.src, args.dst, args.tag);
+ req_storage->remove(request);
if (request == MPI_REQUEST_NULL) {
/* Assume that the trace is well formed, meaning the comm might have been caught by a MPI_test. Then just
class SendAction : public ReplayAction<SendRecvParser> {
public:
SendAction() = delete;
- explicit SendAction(std::string name) : ReplayAction(name) {}
+ explicit SendAction(std::string name, RequestStorage& storage) : ReplayAction(name, storage) {}
void kernel(simgrid::xbt::ReplayAction& action) override
{
int dst_traced = MPI_COMM_WORLD->group()->actor(args.partner)->get_pid();
Request::send(nullptr, args.size, args.datatype1, args.partner, args.tag, MPI_COMM_WORLD);
} else if (name == "Isend") {
MPI_Request request = Request::isend(nullptr, args.size, args.datatype1, args.partner, args.tag, MPI_COMM_WORLD);
- get_reqq_self()->push_back(request);
+ req_storage->add(request);
} else {
xbt_die("Don't know this action, %s", name.c_str());
}
class RecvAction : public ReplayAction<SendRecvParser> {
public:
RecvAction() = delete;
- explicit RecvAction(std::string name) : ReplayAction(name) {}
+ explicit RecvAction(std::string name, RequestStorage& storage) : ReplayAction(name, storage) {}
void kernel(simgrid::xbt::ReplayAction& action) override
{
int src_traced = MPI_COMM_WORLD->group()->actor(args.partner)->get_pid();
Request::recv(nullptr, args.size, args.datatype1, args.partner, args.tag, MPI_COMM_WORLD, &status);
} else if (name == "Irecv") {
MPI_Request request = Request::irecv(nullptr, args.size, args.datatype1, args.partner, args.tag, MPI_COMM_WORLD);
- get_reqq_self()->push_back(request);
+ req_storage->add(request);
}
TRACE_smpi_comm_out(my_proc_id);
class TestAction : public ReplayAction<WaitTestParser> {
public:
- TestAction() : ReplayAction("Test") {}
+ TestAction(RequestStorage& storage) : ReplayAction("Test", storage) {}
void kernel(simgrid::xbt::ReplayAction& action) override
{
- MPI_Request request = get_reqq_self()->back();
- get_reqq_self()->pop_back();
+ MPI_Request request = req_storage->find(args.src, args.dst, args.tag);
+ req_storage->remove(request);
// if request is null here, this may mean that a previous test has succeeded
// Different times in traced application and replayed version may lead to this
// In this case, ignore the extra calls.
- if (request != nullptr) {
+ if (request != MPI_REQUEST_NULL) {
TRACE_smpi_testing_in(my_proc_id);
MPI_Status status;
XBT_DEBUG("MPI_Test result: %d", flag);
/* push back request in vector to be caught by a subsequent wait. if the test did succeed, the request is now
* nullptr.*/
- get_reqq_self()->push_back(request);
+ if (request == MPI_REQUEST_NULL)
+ req_storage->addNullRequest(args.src, args.dst, args.tag);
+ else
+ req_storage->add(request);
TRACE_smpi_testing_out(my_proc_id);
}
class WaitAllAction : public ReplayAction<ActionArgParser> {
public:
- WaitAllAction() : ReplayAction("waitAll") {}
+ WaitAllAction(RequestStorage& storage) : ReplayAction("waitAll", storage) {}
void kernel(simgrid::xbt::ReplayAction& action) override
{
- const unsigned int count_requests = get_reqq_self()->size();
+ const unsigned int count_requests = req_storage->size();
if (count_requests > 0) {
TRACE_smpi_comm_in(my_proc_id, __func__, new simgrid::instr::Pt2PtTIData("waitAll", -1, count_requests, ""));
std::vector<std::pair</*sender*/int,/*recv*/int>> sender_receiver;
- for (const auto& req : (*get_reqq_self())) {
+ std::vector<MPI_Request> reqs;
+ req_storage->get_requests(reqs);
+ for (const auto& req : reqs) {
if (req && (req->flags() & RECV)) {
sender_receiver.push_back({req->src(), req->dst()});
}
}
MPI_Status status[count_requests];
- Request::waitall(count_requests, &(*get_reqq_self())[0], status);
+ Request::waitall(count_requests, &(reqs.data())[0], status);
for (auto& pair : sender_receiver) {
TRACE_smpi_recv(pair.first, pair.second, 0);
} // Replay Namespace
}} // namespace simgrid::smpi
+std::vector<simgrid::smpi::replay::RequestStorage> storage;
/** @brief Only initialize the replay, don't do it for real */
void smpi_replay_init(int* argc, char*** argv)
{
smpi_process()->set_replaying(true);
int my_proc_id = simgrid::s4u::this_actor::get_pid();
+ storage.resize(smpi_process_count());
+
TRACE_smpi_init(my_proc_id);
TRACE_smpi_computing_init(my_proc_id);
TRACE_smpi_comm_in(my_proc_id, "smpi_replay_run_init", new simgrid::instr::NoOpTIData("init"));
xbt_replay_action_register("comm_split",[](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::CommunicatorAction().execute(action); });
xbt_replay_action_register("comm_dup", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::CommunicatorAction().execute(action); });
- xbt_replay_action_register("send", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::SendAction("send").execute(action); });
- xbt_replay_action_register("Isend", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::SendAction("Isend").execute(action); });
- xbt_replay_action_register("recv", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::RecvAction("recv").execute(action); });
- xbt_replay_action_register("Irecv", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::RecvAction("Irecv").execute(action); });
- xbt_replay_action_register("test", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::TestAction().execute(action); });
- xbt_replay_action_register("wait", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::WaitAction().execute(action); });
- xbt_replay_action_register("waitAll", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::WaitAllAction().execute(action); });
+ xbt_replay_action_register("send", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::SendAction("send", storage[simgrid::s4u::this_actor::get_pid()-1]).execute(action); });
+ xbt_replay_action_register("Isend", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::SendAction("Isend", storage[simgrid::s4u::this_actor::get_pid()-1]).execute(action); });
+ xbt_replay_action_register("recv", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::RecvAction("recv", storage[simgrid::s4u::this_actor::get_pid()-1]).execute(action); });
+ xbt_replay_action_register("Irecv", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::RecvAction("Irecv", storage[simgrid::s4u::this_actor::get_pid()-1]).execute(action); });
+ xbt_replay_action_register("test", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::TestAction(storage[simgrid::s4u::this_actor::get_pid()-1]).execute(action); });
+ xbt_replay_action_register("wait", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::WaitAction(storage[simgrid::s4u::this_actor::get_pid()-1]).execute(action); });
+ xbt_replay_action_register("waitAll", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::WaitAllAction(storage[simgrid::s4u::this_actor::get_pid()-1]).execute(action); });
xbt_replay_action_register("barrier", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::BarrierAction().execute(action); });
xbt_replay_action_register("bcast", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::BcastAction().execute(action); });
xbt_replay_action_register("reduce", [](simgrid::xbt::ReplayAction& action) { simgrid::smpi::replay::ReduceAction().execute(action); });
