graph_helper.cc 22.2 KB
Newer Older
X
better  
Xin Pan 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14
/* Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserved.

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License. */

C
chengduo 已提交
15
#include "paddle/fluid/framework/ir/graph_helper.h"
16

17
#include <queue>
Y
Yan Chunwei 已提交
18
#include <stack>
19

20
#include "paddle/fluid/framework/details/multi_devices_helper.h"
21
#include "paddle/fluid/framework/op_proto_maker.h"
X
better  
Xin Pan 已提交
22

23
DECLARE_bool(convert_all_blocks);
24 25
PADDLE_DEFINE_EXPORTED_string(print_sub_graph_dir,
                              "",
26 27
                              "FLAGS_print_sub_graph_dir is used "
                              "to print the nodes of sub_graphs.");
C
chengduo 已提交
28

X
better  
Xin Pan 已提交
29 30 31 32
namespace paddle {
namespace framework {
namespace ir {
namespace {
33 34

template <class NodeComparator = ir::NodeComp>
35 36
void SortHelper(const std::map<ir::Node *,
                               std::set<ir::Node *, NodeComparator>,
37
                               NodeComparator> &adj_list,
38 39
                ir::Node *node,
                std::unordered_set<ir::Node *> *visited,
40
                std::vector<ir::Node *> *ret) {
X
better  
Xin Pan 已提交
41 42 43 44
  visited->insert(node);

  for (auto adj : adj_list.at(node)) {
    if (visited->find(adj) == visited->end()) {
45
      SortHelper<NodeComparator>(adj_list, adj, visited, ret);
X
better  
Xin Pan 已提交
46 47 48
    }
  }

Y
Yan Chunwei 已提交
49
  VLOG(5) << "topology sort insert: " << node->Name() << " "
M
minqiyang 已提交
50
          << reinterpret_cast<void *>(node) << " input " << node->inputs.size();
X
better  
Xin Pan 已提交
51 52 53
  ret->push_back(node);
}

54
template <class NodeComparator = ir::NodeComp>
55 56 57 58 59 60 61
bool HasCircleHelper(ir::Node *node,
                     const std::map<ir::Node *,
                                    std::set<ir::Node *, NodeComparator>,
                                    NodeComparator> &adj_list,
                     std::unordered_set<ir::Node *> *visited,
                     std::unordered_set<ir::Node *> *in_trace,
                     std::vector<std::vector<ir::Node *>> *circles) {
X
better  
Xin Pan 已提交
62 63 64 65 66 67
  if (visited->find(node) == visited->end()) {
    visited->insert(node);
    in_trace->insert(node);

    for (ir::Node *in : adj_list.at(node)) {
      if (visited->find(in) == visited->end() &&
68 69
          HasCircleHelper<NodeComparator>(
              in, adj_list, visited, in_trace, circles)) {
X
better  
Xin Pan 已提交
70 71
        return true;
      } else if (in_trace->find(in) != in_trace->end()) {
D
dzhwinter 已提交
72 73 74 75 76 77 78 79 80 81 82 83
        if (circles != nullptr) {
          std::vector<ir::Node *> circle;
          circle.emplace_back(in);
          ir::Node *p = in;
          for (auto &adj : adj_list.at(p)) {
            if (in_trace->count(adj)) {
              circle.emplace_back(adj);
              p = adj;
            }
          }
          circles->emplace_back(circle);
        }
X
better  
Xin Pan 已提交
84 85 86 87 88 89 90 91
        return true;
      }
    }
  }
  in_trace->erase(node);
  return false;
}

92
template <class NodeComparator = ir::NodeComp>
93 94 95 96
bool HasCircleInternal(const std::map<ir::Node *,
                                      std::set<ir::Node *, NodeComparator>,
                                      NodeComparator> &adj_list,
                       std::vector<std::vector<ir::Node *>> *circles) {
X
better  
Xin Pan 已提交
97 98 99
  std::unordered_set<ir::Node *> visited;
  std::unordered_set<ir::Node *> in_trace;
  for (auto &adj : adj_list) {
100 101
    if (HasCircleHelper<NodeComparator>(
            adj.first, adj_list, &visited, &in_trace, circles)) {
X
better  
Xin Pan 已提交
102 103 104 105 106
      return true;
    }
  }
  return false;
}
X
Xin Pan 已提交
107 108 109
}  // namespace

bool HasCircle(const Graph &graph) {
D
dzhwinter 已提交
110 111 112
  return HasCircleInternal(BuildOperationAdjList(graph), nullptr);
}

113 114 115 116 117 118 119 120 121 122
bool VarDescIsConsistency(const Graph &graph) {
  std::unordered_map<std::string, std::unordered_set<ir::Node *>>
      var_name2node_set;
  for (ir::Node *node : graph.Nodes()) {
    if (node->IsVar() && node->Var()) {
      var_name2node_set[node->Var()->Name()].emplace(node);
    }
  }
  for (auto &iter : var_name2node_set) {
    auto &first_node = *iter.second.begin();
123 124
    bool is_persistable = std::any_of(iter.second.begin(),
                                      iter.second.end(),
125 126 127 128 129
                                      [&first_node](const ir::Node *node) {
                                        return node->Var()->Persistable();
                                      });
    if (is_persistable) {
      bool is_consistency =
130 131
          std::all_of(iter.second.begin(),
                      iter.second.end(),
132 133 134 135 136 137 138 139
                      [&first_node](const ir::Node *node) {
                        return *node->Var() == *first_node->Var();
                      });
      if (!is_consistency) return false;
    }
  }
  return true;
}
D
dzhwinter 已提交
140 141 142
bool FindCircleSubGraph(const Graph &graph,
                        std::vector<std::vector<ir::Node *>> *circles) {
  return HasCircleInternal(BuildOperationAdjList(graph), circles);
X
Xin Pan 已提交
143
}
X
better  
Xin Pan 已提交
144

X
Xin Pan 已提交
145
std::vector<ir::Node *> TopologySortOperations(const Graph &graph) {
146 147
  std::map<ir::Node *, std::set<ir::Node *, ir::NodeComp>, ir::NodeComp>
      adj_list = BuildOperationAdjList(graph);
148 149
  PADDLE_ENFORCE_EQ(HasCircleInternal(adj_list, nullptr),
                    false,
150 151
                    platform::errors::InvalidArgument(
                        "Generated graph shouldn't contain cycle."));
X
better  
Xin Pan 已提交
152 153 154 155
  std::unordered_set<ir::Node *> visited;
  std::vector<ir::Node *> ret;
  for (auto adj : adj_list) {
    if (visited.find(adj.first) == visited.end()) {
156
      SortHelper<ir::NodeComp>(adj_list, adj.first, &visited, &ret);
X
better  
Xin Pan 已提交
157 158
    }
  }
159

X
better  
Xin Pan 已提交
160 161 162
  return ret;
}

Z
Zeng Jinle 已提交
163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188
bool IsTopologySortOperationsUnique(const Graph &graph) {
  auto nodes = TopologySortOperations(graph);
  size_t n = nodes.size();
  for (size_t i = 1; i < n; ++i) {
    auto *prev_op = nodes[i - 1];
    auto *cur_op = nodes[i];

    std::unordered_set<Node *> prev_op_outputs;
    for (auto *output : prev_op->outputs) {
      prev_op_outputs.insert(output);
    }

    bool found = false;
    for (auto *input : cur_op->inputs) {
      if (prev_op_outputs.count(input) > 0) {
        found = true;
        break;
      }
    }
    if (!found) {
      return false;
    }
  }
  return true;
}

Y
Yan Chunwei 已提交
189 190 191 192 193 194 195 196 197 198 199 200
// Build operator outlink edge table.
std::map<ir::Node *, std::unordered_set<ir::Node *>> BuildOperationOutAdjList(
    const Graph &graph) {
  std::map<ir::Node *, std::unordered_set<ir::Node *>> adj_list;

  for (auto &n : graph.Nodes()) {
    if (!n->IsOp()) continue;
    if (adj_list.find(n) == adj_list.end()) {
      adj_list[n] = std::unordered_set<ir::Node *>();
    }
    for (auto &var : n->outputs) {
      for (auto &adj_n : var->outputs) {
201 202 203 204 205 206
        PADDLE_ENFORCE_EQ(adj_n->NodeType(),
                          ir::Node::Type::kOperation,
                          platform::errors::InvalidArgument(
                              "Node(%s)'s type(%d) must be kOperation type.",
                              adj_n->Name(),
                              static_cast<int>(adj_n->NodeType())));
Y
Yan Chunwei 已提交
207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306
        VLOG(40) << "adj " << adj_n->Name() << reinterpret_cast<void *>(adj_n)
                 << " -> " << n->Name() << reinterpret_cast<void *>(n)
                 << "  via " << var->Name() << reinterpret_cast<void *>(var);
        adj_list[n].insert(adj_n);
      }
    }
  }
  return adj_list;
}

std::vector<ir::Node *> OpDFSSort(const Graph &graph) {
  auto edge_table = BuildOperationOutAdjList(graph);
  std::stack<Node *> stack;
  for (auto &ele : edge_table) {
    if (ele.first->inputs.empty()) {
      // find the input ops (those without input vars)
      stack.push(ele.first);
    } else {
      // find the ops with only persistable vars as inputs.
      bool all_persistable = true;
      for (auto *input : ele.first->inputs) {
        if (!(input->IsVar() && input->Var() && input->Var()->Persistable())) {
          all_persistable = false;
        }
      }
      if (all_persistable) {
        stack.push(ele.first);
      }
    }
  }

  std::vector<Node *> res;
  // start from the feed op and DFS
  std::unordered_set<Node *> unique_set;
  while (!stack.empty()) {
    // will start from the last feed by default.
    auto cur = stack.top();
    stack.pop();
    unique_set.insert(cur);
    res.push_back(cur);

    for (auto *op : edge_table[cur]) {
      if (!unique_set.count(op)) {
        stack.push(op);
      }
    }
  }
  return res;
}

std::vector<ir::Node *> TopologyDfsSortOperations(const Graph &graph) {
  std::vector<ir::Node *> nodes;
  std::unordered_map<Node *, int> in_degree;

  auto set_out_ops_ready = [&](Node *var) {
    for (auto *op : var->outputs) {
      --in_degree[op];
    }
  };
  // build in_degree
  for (auto *node : graph.Nodes()) {
    if (node->IsOp()) {
      in_degree[node] += node->inputs.size();
    } else if (node->IsVar() && node->inputs.empty()) {
      // put all the inputs of the whole graph ready.
      set_out_ops_ready(node);
    }
  }

  std::deque<Node *> op_queue;
  // first visit
  for (auto &node : OpDFSSort(graph)) {
    if (node->IsOp()) {
      op_queue.push_back(node);
    }
  }

  // traverse the graph
  int num_ops = op_queue.size();
  while (num_ops) {
    for (auto it = op_queue.begin(); it != op_queue.end(); it++) {
      auto *&cur_op = *it;
      if (!cur_op || in_degree[cur_op] > 0) continue;
      // visit this node
      // put all the output var of this op valid.
      for (auto *out_var : cur_op->outputs) {
        if (!out_var) continue;
        set_out_ops_ready(out_var);
      }
      VLOG(8) << "visit " << cur_op->Name();
      nodes.push_back(cur_op);

      cur_op = nullptr;
      num_ops--;
    }
  }

  return nodes;
}

C
chengduo 已提交
307
size_t GraphNum(const Graph &graph) {
D
dzhwinter 已提交
308
  std::unordered_set<ir::Node *> nodes(graph.Nodes());
C
chengduo 已提交
309 310 311 312 313
  std::unordered_set<ir::Node *> visited_nodes;
  visited_nodes.reserve(nodes.size());
  std::deque<ir::Node *> q_nodes;
  std::vector<std::unordered_set<ir::Node *>> graph_nodes;
  std::unordered_set<ir::Node *> g_nodes;
W
Wu Yi 已提交
314 315
  // q_set used to record records in the queue.
  std::unordered_set<ir::Node *> q_set;
C
chengduo 已提交
316 317
  size_t graph_count = 0;

318 319 320 321 322 323 324 325 326
  auto traverse_nodes =
      [&visited_nodes, &q_nodes, &q_set](const std::vector<ir::Node *> &nodes) {
        for (auto n : nodes) {
          if (visited_nodes.count(n) == 0 && q_set.count(n) == 0) {
            q_nodes.push_back(n);
            q_set.insert(n);
          }
        }
      };
C
chengduo 已提交
327 328 329 330 331

  while (visited_nodes.size() != nodes.size()) {
    if (!q_nodes.empty()) {
      auto cur_node = q_nodes.front();
      q_nodes.pop_front();
W
Wu Yi 已提交
332
      q_set.erase(cur_node);
C
chengduo 已提交
333 334 335 336 337 338 339 340 341 342 343 344 345
      visited_nodes.insert(cur_node);
      g_nodes.insert(cur_node);
      traverse_nodes(cur_node->inputs);
      traverse_nodes(cur_node->outputs);
    } else {
      ++graph_count;
      if (g_nodes.size()) {
        graph_nodes.emplace_back(g_nodes);
      }
      g_nodes.clear();
      for (auto &n : nodes) {
        if (visited_nodes.count(n) == 0) {
          q_nodes.push_back(n);
W
Wu Yi 已提交
346
          q_set.insert(n);
C
chengduo 已提交
347 348 349 350 351 352 353 354 355 356
          break;
        }
      }
    }
  }

  if (g_nodes.size()) {
    graph_nodes.emplace_back(g_nodes);
  }

C
chengduo 已提交
357 358 359 360 361 362 363 364 365
  if (FLAGS_print_sub_graph_dir.size()) {
    if (graph_nodes.size() > 1) {
      std::stringstream out;
      for (auto &g_n : graph_nodes) {
        out << "graph_nodes: " << g_n.size() << "\n";
      }
      out << "\n\n";
      for (auto &g_n : graph_nodes) {
        out << "graph_nodes: " << g_n.size();
C
chengduo 已提交
366 367 368 369 370 371 372 373 374 375 376
        for (auto &node : g_n) {
          out << "\nNode: " << node->Name() << " in [";
          for (auto &n : node->inputs) {
            out << n->Name() << ", ";
          }
          out << "], out[";
          for (auto &n : node->outputs) {
            out << n->Name() << ", ";
          }
          out << "]";
        }
C
chengduo 已提交
377
        out << "\n\n\n";
C
chengduo 已提交
378
      }
C
chengduo 已提交
379 380
      std::unique_ptr<std::ostream> fout(
          new std::ofstream(FLAGS_print_sub_graph_dir));
381 382
      PADDLE_ENFORCE_EQ(fout->good(),
                        true,
383 384 385
                        platform::errors::Unavailable(
                            "Can not open file %s for printing the graph.",
                            FLAGS_print_sub_graph_dir));
C
chengduo 已提交
386
      *fout << out.str();
C
chengduo 已提交
387 388 389 390 391 392
    }
  }

  return graph_count;
}

Y
Yan Chunwei 已提交
393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415
void CleanIndividualNodes(Graph *graph) {
  std::unordered_set<Node *> nodes2rm;
  for (auto *node : graph->Nodes()) {
    if (node->inputs.empty() && node->outputs.empty()) {
      nodes2rm.insert(node);
    }
  }

  for (auto *node : nodes2rm) {
    graph->RemoveNode(node);
  }
}

std::vector<Node *> TopologyVarientSort(const Graph &graph,
                                        SortKind sort_kind) {
  switch (sort_kind) {
    case SortKind::TS:
      return framework::ir::TopologySortOperations(graph);
    default:
      return framework::ir::TopologyDfsSortOperations(graph);
  }
}

416 417
class DescOrderComparator {
 public:
418 419 420 421 422 423 424 425
  bool operator()(Node *const &n1, Node *const &n2) const {
    if (n1->DescOrder() < n2->DescOrder()) {
      return true;
    } else if (n1->DescOrder() == n2->DescOrder()) {
      return n1->id() < n2->id() ||
             (n1->id() == n2->id() && n1->ToString() < n2->ToString());
    }
    return false;
426 427 428 429
  }
};

std::vector<ir::Node *> TopologySortGraphByDescOrder(const Graph &graph) {
430 431
  std::map<ir::Node *,
           std::set<ir::Node *, DescOrderComparator>,
432 433 434
           DescOrderComparator>
      adj_list = BuildOperationAdjList<DescOrderComparator>(graph);
  PADDLE_ENFORCE_EQ(HasCircleInternal<DescOrderComparator>(adj_list, nullptr),
435 436 437
                    false,
                    platform::errors::InvalidArgument(
                        "Generated graph shouldn't contain cycle."));
438 439 440 441 442
  std::unordered_set<ir::Node *> visited;
  std::vector<ir::Node *> ret;
  for (auto adj : adj_list) {
    if (visited.find(adj.first) == visited.end()) {
      SortHelper<DescOrderComparator>(adj_list, adj.first, &visited, &ret);
443 444 445
    }
  }

446
  return ret;
447 448
}

449 450 451 452 453 454
static OpDesc *ReplaceScaleLossGradOp(const Node &node, OpDesc *desc) {
  desc->SetType("fill_constant");
  desc->SetAttr(
      OpProtoAndCheckerMaker::OpRoleAttrName(),
      (static_cast<int>(OpRole::kBackward) | static_cast<int>(OpRole::kLoss)));
  desc->SetAttr("value", 1.0f);
455
  desc->SetAttr("shape", std::vector<int64_t>({1}));
456 457 458 459 460 461 462 463 464 465
  std::vector<std::string> output_names;
  for (auto out : node.outputs) {
    output_names.emplace_back(out->Name());
  }
  desc->SetOutput("Out", output_names);
  return desc;
}

static void GetGraphOpDesc(const std::vector<Node *> &nodes,
                           std::vector<OpDesc> *ops) {
466 467 468 469 470 471 472 473 474 475 476 477 478
  auto is_fused_opt = [](Node *n) -> bool {
    auto op_type = n->Op()->Type();
    auto is_opt =
        (op_type == "adam" || op_type == "momentum" || op_type == "sgd");
    auto input_names = n->Op()->InputArgumentNames();
    auto contains_fused_var = std::any_of(
        input_names.begin(), input_names.end(), [](std::string name) {
          return name.find(details::kFusedVarNamePrefix) != std::string::npos;
        });
    VLOG(4) << is_opt << " " << contains_fused_var;
    return is_opt && contains_fused_var;
  };

479 480 481 482 483 484 485 486 487 488
  for (Node *n : nodes) {
    // if node is not Op, skip
    if (!n->IsOp()) continue;

    // create fill_constant op
    if (n->Name() == "scale_loss_grad") {
      ops->emplace_back();
      auto &desc = ops->back();
      ReplaceScaleLossGradOp(*n, &desc);
    } else if (n->Op()) {
489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508
      VLOG(4) << "convert op node to desc " << n->Op()->Type();
      VLOG(4) << n->ToString();
      if (is_fused_opt(n)) {
        OpDesc depend_desc(n->Op()->Block());

        std::vector<std::string> deps;
        for (auto in : n->inputs) {
          if (in->IsVar() && !in->IsCtrlVar()) {
            deps.push_back(in->Name());
          }
        }
        depend_desc.SetType("depend");
        depend_desc.SetInput("X",
                             n->Op()->Inputs().at(n->Op()->InputNames()[0]));
        depend_desc.SetInput("Dep", deps);
        depend_desc.SetOutput("Out",
                              n->Op()->Inputs().at(n->Op()->InputNames()[0]));
        ops->emplace_back(depend_desc);
        VLOG(4) << "add depend op";
      }
509 510 511 512 513 514
      ops->emplace_back(*n->Op());
    }
    // delete no OpDesc op
  }
}

515 516
static void GraphToBlock(const Graph &graph,
                         proto::BlockDesc *block,
517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559
                         const SortKind *sort_kind) {
  // Remove the unneeded variables after memory optimization.
  std::unordered_set<std::string> vars2remove;
  if (graph.Has(kGraphToProgramVarsToRemove)) {
    vars2remove =
        graph.Get<std::unordered_set<std::string>>(kGraphToProgramVarsToRemove);
    VLOG(2) << "graph (id: " << block->idx() << ") to program remove "
            << vars2remove.size() << " nodes";
  }

  block->clear_vars();
  std::unordered_set<std::string> visited_vars;
  for (Node *n : graph.Nodes()) {
    if (n->IsVar()) {
      if (n->Var() && visited_vars.count(n->Var()->Name()) == 0 &&
          !vars2remove.count(n->Var()->Name()) &&
          n->GetVarNodeBlockId() == graph.GetBlockId()) {
        visited_vars.insert(n->Var()->Name());
        block->add_vars()->MergeFrom(*n->Var()->Proto());
      }
    }
  }
  block->clear_ops();

  std::vector<Node *> nodes;
  if (sort_kind != nullptr) {
    // Inference Memory Optimize relays on this branch.
    nodes = TopologyVarientSort(graph, *sort_kind);
  } else {
    if (FLAGS_convert_all_blocks) {
      nodes = TopologySortGraphByDescOrder(graph);
    } else {
      nodes = TopologySortOperations(graph);
    }
  }

  std::vector<OpDesc> ops;
  GetGraphOpDesc(nodes, &ops);
  for (auto &op : ops) {
    block->add_ops()->MergeFrom(*op.Proto());
  }
}

560 561
void GraphToProgram(const Graph &graph,
                    ProgramDesc *program,
562
                    const SortKind *sort_kind) {
563 564
  PADDLE_ENFORCE_EQ(graph.IsMainGraph(),
                    true,
565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587
                    platform::errors::InvalidArgument(
                        "This graph is a sub_graph, "
                        "and can't convert to program individually"));
  PADDLE_ENFORCE_NOT_NULL(
      program,
      platform::errors::InvalidArgument(
          "program must not be nullptr when converting graph to program"));

  proto::ProgramDesc program_pb(*(program->Proto()));
  auto block = program_pb.mutable_blocks(kRootBlockIndex);
  block->set_idx(kRootBlockIndex);

  if (FLAGS_convert_all_blocks) {
    GraphToBlock(*graph.GetSubGraph(kRootBlockIndex), block, sort_kind);

    VLOG(3) << "Graph to program need convert " << graph.SubGraphsSize()
            << " sub graph";
    for (size_t idx = 0; idx < graph.SubGraphsSize(); ++idx) {
      // avoid kRootBlockIndex not 0
      if (idx == kRootBlockIndex) continue;

      block = program_pb.add_blocks();
      block->set_idx(idx);
588
      block->set_parent_idx(kRootBlockIndex);
589 590 591 592 593 594 595 596 597
      GraphToBlock(*graph.GetSubGraph(idx), block, sort_kind);
    }
  } else {
    GraphToBlock(graph, block, sort_kind);
  }

  program->CopyFrom(program_pb);
}

598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655
static std::vector<std::vector<ir::Node::Dep>> GetOpDependencies(
    const BlockDesc &block, const std::unordered_set<ir::Node *> &nodes) {
  auto block_ops = block.AllOps();
  size_t op_num = block_ops.size();
  std::unordered_map<const ir::Node *, std::unordered_set<const ir::Node *>>
      preceding_ops(op_num);
  std::unordered_map<const ir::Node *, size_t> preceding_deps(op_num);
  std::unordered_map<const ir::Node *, std::unordered_set<const ir::Node *>>
      pending_ops(op_num);

  std::queue<const ir::Node *> ready_ops;
  for (const auto *node : nodes) {
    if (!node->IsOp()) continue;

    auto &tmp_preceding_ops = preceding_ops[node];
    for (const auto *in_var : node->inputs) {
      for (const auto *in_op : in_var->inputs) {
        tmp_preceding_ops.insert(in_op);
      }
    }
    if (tmp_preceding_ops.empty()) {
      ready_ops.push(node);
    }
    preceding_deps[node] = tmp_preceding_ops.size();

    auto &tmp_pending_ops = pending_ops[node];
    for (const auto *out_var : node->outputs) {
      for (const auto *out_op : out_var->outputs) {
        tmp_pending_ops.insert(out_op);
      }
    }
  }

  std::unordered_map<const ir::Node *, std::unordered_set<const ir::Node *>>
      all_preceding_ops;
  while (!ready_ops.empty()) {
    const auto *cur_op = ready_ops.front();
    ready_ops.pop();

    auto &all_preceding_ops_of_cur_op = all_preceding_ops[cur_op];
    for (const auto *preceding_op : preceding_ops.at(cur_op)) {
      all_preceding_ops_of_cur_op.insert(preceding_op);
      auto &prev_preceding_ops = all_preceding_ops[preceding_op];
      all_preceding_ops_of_cur_op.insert(prev_preceding_ops.begin(),
                                         prev_preceding_ops.end());
    }

    for (const auto *pending_op : pending_ops.at(cur_op)) {
      if (--preceding_deps.at(pending_op) == 0) {
        ready_ops.push(pending_op);
      }
    }
  }

  std::unordered_map<uint64_t, size_t> op_id_to_idx(op_num);
  for (const auto *op_desc : block_ops) {
    size_t op_idx = op_id_to_idx.size();
    PADDLE_ENFORCE_EQ(
656 657
        op_id_to_idx.emplace(op_desc->OriginalId(), op_idx).second,
        true,
658
        platform::errors::InvalidArgument(
S
sneaxiy 已提交
659
            "There should not be duplicate op id: %d", op_desc->OriginalId()));
660 661 662 663 664 665 666 667 668 669
  }

  std::vector<std::vector<ir::Node::Dep>> dep_matrix(op_num);
  for (size_t i = 0; i < op_num; ++i) {
    dep_matrix[i].resize(op_num, ir::Node::Dep::kNoDep);
    dep_matrix[i][i] = ir::Node::Dep::kSame;
  }

  auto get_op_idx_by_id = [&op_id_to_idx](uint64_t op_id) {
    auto iter = op_id_to_idx.find(op_id);
670 671
    PADDLE_ENFORCE_NE(iter,
                      op_id_to_idx.end(),
672 673 674 675 676 677 678
                      platform::errors::InvalidArgument(
                          "Cannot find OpDesc with id %d", op_id));
    return iter->second;
  };

  for (const auto &pair : all_preceding_ops) {
    const auto *cur_op_node = pair.first;
S
sneaxiy 已提交
679
    size_t op_idx_1 = get_op_idx_by_id(cur_op_node->Op()->OriginalId());
680
    for (const auto *preceding_op_node : pair.second) {
S
sneaxiy 已提交
681
      size_t op_idx_2 = get_op_idx_by_id(preceding_op_node->Op()->OriginalId());
682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701
      dep_matrix[op_idx_1][op_idx_2] = ir::Node::Dep::kAfter;
      dep_matrix[op_idx_2][op_idx_1] = ir::Node::Dep::kBefore;
    }
  }
  return dep_matrix;
}

std::vector<std::vector<std::vector<ir::Node::Dep>>> GetOpDependencies(
    const ProgramDesc &program) {
  ir::Graph graph(program);
  size_t block_num = program.Size();
  std::vector<std::vector<std::vector<ir::Node::Dep>>> deps;
  deps.reserve(block_num);
  for (size_t i = 0; i < block_num; ++i) {
    deps.emplace_back(
        GetOpDependencies(program.Block(i), graph.GetSubGraph(i)->Nodes()));
  }
  return deps;
}

X
better  
Xin Pan 已提交
702 703 704
}  // namespace ir
}  // namespace framework
}  // namespace paddle