memory_optimize_helper.cc 14.3 KB
Newer Older
D
dzhwinter 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.

D
dzhwinter 已提交
15
#include "paddle/fluid/framework/details/memory_optimize_helper.h"
D
dzhwinter 已提交
16
#include <deque>
D
dzhwinter 已提交
17
#include <functional>
D
dzhwinter 已提交
18
#include <iostream>
D
dzhwinter 已提交
19
#include <numeric>
D
dzhwinter 已提交
20 21
#include <sstream>
#include <string>
D
dzhwinter 已提交
22
#include "paddle/fluid/framework/var_desc.h"
D
dzhwinter 已提交
23 24 25 26

namespace paddle {
namespace framework {
namespace details {
D
dzhwinter 已提交
27
using paddle::framework::VarDesc;
D
dzhwinter 已提交
28

D
dzhwinter 已提交
29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117
std::vector<ir::Node*> SortOpLikeDescOrder(const ir::Graph& graph) {
  PADDLE_ENFORCE(graph.Has(kAllOpDescs),
                 "Graph has no attribute of kAllOpDescs.");
  // 1. get op desc order
  auto& op_descs = graph.Get<const std::vector<OpDesc*>>(kAllOpDescs);

  // 2. topology sort order
  auto nodes = graph.Nodes();
  std::deque<ir::Node*> ops;
  FilterVariables(nodes, [&](ir::Node* op) {
    if (op->IsOp() && op->Op() != nullptr) {
      ops.emplace_back(op);
    }
  });
  std::unordered_map<ir::Node*, size_t> op_deps;
  std::list<ir::Node*> ready_ops;
  std::unordered_map<ir::Node*, std::unordered_set<ir::Node*>> pending_ops;

  for (auto* op : ops) {
    std::unordered_set<ir::Node*> preceding_op;
    for (auto* in : op->inputs) {
      if (in->inputs.empty()) continue;
      PADDLE_ENFORCE(in->inputs.size() == 1 && in->inputs[0]->IsOp());
      preceding_op.emplace(in->inputs[0]);
      pending_ops[in->inputs[0]].emplace(op);
    }
    op_deps[op] = preceding_op.size();
    if (preceding_op.empty()) {
      ready_ops.emplace_back(op);
    }
  }

  // 3. generated op list based desc order and the topology order
  std::vector<ir::Node*> ret;
  std::list<OpDesc*> op_descs_list(op_descs.begin(), op_descs.end());

  auto update_by_found_node = [&](ir::Node* found_node) {
    for (auto* pending_op : pending_ops[found_node]) {
      if (--op_deps[pending_op] == 0) {
        ready_ops.emplace_back(pending_op);
      }
    }
    ready_ops.remove(found_node);
    ret.emplace_back(found_node);
  };

  while (!ready_ops.empty()) {
    bool all_of_ready_op_unmatched = true;
    for (auto it = op_descs_list.begin(); it != op_descs_list.end();) {
      auto op_desc = *it;
      ir::Node* found_node = nullptr;
      for (auto* op : ready_ops) {
        if (IsSameDesc(op->Op(), op_desc)) {
          found_node = op;
          break;
        }
      }

      // 3.1 op desc deleted by other pass
      if (found_node == nullptr) {
        ++it;
        continue;
      } else {
        all_of_ready_op_unmatched = false;
        it = op_descs_list.erase(it);
      }
      update_by_found_node(found_node);
    }

    // 3.2 op descs are added by other pass
    // preceding op non empty means some new op descs are
    // created, but not contained in return node list.
    // these new op desc may depend on each other.
    std::list<ir::Node*> prev_ready_ops(ready_ops);
    if (all_of_ready_op_unmatched) {
      for (auto op : prev_ready_ops) {
        update_by_found_node(op);
      }
    }
  }

  PADDLE_ENFORCE(std::all_of(
      op_deps.begin(), op_deps.end(),
      [&](const std::pair<ir::Node*, size_t>& p) { return p.second == 0; }));

  return ret;
}

size_t NodeSize(const VarDesc& node) {
D
dzhwinter 已提交
118 119 120 121 122 123 124
  auto shape = node.GetShape();
  int size =
      std::accumulate(shape.begin(), shape.end(), 1, std::multiplies<int>());
  size_t type_size = SizeOfType(node.GetDataType());
  return type_size * std::abs(size);
}

D
dzhwinter 已提交
125
size_t NodeSize(ir::Node* n) {
D
dzhwinter 已提交
126
  auto* desc = FindVarDescInBlock(n);
D
dzhwinter 已提交
127
  return NodeSize(*desc);
D
dzhwinter 已提交
128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172
}

std::string DebugStringImpl(VarDesc* var) {
  std::stringstream ss;
  ss << var->Name();
  ss << "[";
  try {
    auto shape = var->GetShape();
    for (size_t i = 0; i < shape.size(); ++i) {
      if (i != shape.size() - 1) {
        ss << shape[i] << ",";
      } else {
        ss << shape[i];
      }
    }
    ss << "]";
  } catch (...) {
    ss << "Var has no VarDesc !!! Name:" << var->Name();
  }
  return ss.str();
}

std::string DebugString(ir::Node* var) {
  return DebugStringImpl(FindVarDescInBlock(var));
}

// NOTE(dzh): based ir node, if a large node has been reused
// by a small size node, then next time it appear in pool, it will
// have the small size. Find the original node shap from blockdesc.
VarDesc* FindVarDescInBlock(ir::Node* n) {
  PADDLE_ENFORCE(n->IsVar() && !n->IsCtrlVar() && n->inputs.size() == 1);
  BlockDesc* block = n->inputs[0]->Op()->Block();
  PADDLE_ENFORCE(block->HasVar(n->Name()),
                 string::Sprintf("Block do not has var %s", n->Name()));
  return block->FindVar(n->Name());
}

struct NodeComparator {
  bool operator()(ir::Node* lhs, ir::Node* rhs) const {
    auto* lhs_desc = FindVarDescInBlock(lhs);
    auto* rhs_desc = FindVarDescInBlock(rhs);
    auto lhs_shape = lhs_desc->GetShape();
    auto rhs_shape = rhs_desc->GetShape();
    if ((lhs_shape[0] == -1 && rhs_shape[0] == -1) ||
        (lhs_shape[0] != -1 && rhs_shape[0] != -1)) {
D
dzhwinter 已提交
173
      return NodeSize(lhs) <= NodeSize(rhs);
D
dzhwinter 已提交
174 175 176 177 178 179
    } else {
      return false;
    }
  }
};

D
dzhwinter 已提交
180
void OrderedSet::Insert(ir::Node* var) {
D
dzhwinter 已提交
181 182
  PADDLE_ENFORCE(var->IsVar() && !var->IsCtrlVar());
  if (mark_table_.count(var->Name()) != 0) {
D
dzhwinter 已提交
183
    mark_table_[var->Name()]->emplace_back(var);
D
dzhwinter 已提交
184 185 186 187 188 189 190
    return;
  }

  auto* var_desc = FindVarDescInBlock(var);
  auto var_shape = var_desc->GetShape();
  int batch_size = static_cast<int>(var_shape[0]);

D
dzhwinter 已提交
191
  NodeComparator functor;
D
dzhwinter 已提交
192 193
  Iter it = nodes_.begin();
  while (it != nodes_.end()) {
D
dzhwinter 已提交
194 195
    auto& prev = it->front();
    auto* cache_desc = FindVarDescInBlock(prev);
D
dzhwinter 已提交
196 197 198
    int cache_batch_size = cache_desc->GetShape()[0];
    if ((cache_batch_size == -1 && batch_size == -1) ||
        (cache_batch_size != -1 && batch_size != -1)) {
D
dzhwinter 已提交
199
      if (functor(prev, var)) {
D
dzhwinter 已提交
200 201 202 203 204 205 206 207 208 209 210
        ++it;
      } else {
        break;
      }
    } else if (cache_batch_size == -1 && batch_size != -1) {
      ++it;
    } else if (cache_batch_size != -1 && batch_size == -1) {
      break;
    }
  }

D
dzhwinter 已提交
211
  it = nodes_.insert(it, {var});
D
dzhwinter 已提交
212 213 214
  mark_table_[var->Name()] = it;
}

D
dzhwinter 已提交
215
int OrderedSet::GetNodeIndexInPool(ir::Node* var) {
D
dzhwinter 已提交
216 217 218
  return std::distance(nodes_.begin(), mark_table_[var->Name()]);
}

D
dzhwinter 已提交
219
ir::Node* OrderedSet::FindBestFitNode(ir::Node* var) const {
D
dzhwinter 已提交
220
  ir::Node* found_node = nullptr;
D
dzhwinter 已提交
221
  NodeComparator functor;
D
dzhwinter 已提交
222 223

  for (auto it = nodes_.begin(); it != nodes_.end(); ++it) {
D
dzhwinter 已提交
224 225 226
    auto& candidate = it->front();
    if (functor(var, candidate)) {
      found_node = candidate;
D
dzhwinter 已提交
227 228 229 230 231 232
      break;
    }
  }
  return found_node;
}

D
dzhwinter 已提交
233 234 235 236 237 238 239 240 241 242
bool OrderedSet::Has(ir::Node* var) const {
  if (mark_table_.count(var->Name())) {
    auto& node_in_samename = mark_table_.at(var->Name());
    auto iter =
        std::find_if(node_in_samename->begin(), node_in_samename->end(),
                     [&](ir::Node* n) { return n->Name() == var->Name(); });
    return iter != node_in_samename->end();
  }
  return false;
}
D
dzhwinter 已提交
243

D
dzhwinter 已提交
244 245 246 247
void OrderedSet::Erase(ir::Node* var) {
  PADDLE_ENFORCE(mark_table_.count(var->Name()));
  nodes_.erase(mark_table_[var->Name()]);
  mark_table_.erase(var->Name());
D
dzhwinter 已提交
248 249
}

D
dzhwinter 已提交
250
std::string OrderedSet::ToString() const {
D
dzhwinter 已提交
251 252
  std::stringstream ss;
  for (auto it = nodes_.begin(); it != nodes_.end(); ++it) {
D
dzhwinter 已提交
253 254 255
    for (auto& node : *it) {
      ss << DebugString(node) << " ";
    }
D
dzhwinter 已提交
256 257 258 259
  }
  return ss.str();
}

D
dzhwinter 已提交
260
bool NodeCanReused(ir::Node* node) {
D
dzhwinter 已提交
261
  // valid the node is a var node
D
dzhwinter 已提交
262
  if (node == nullptr || !node->IsVar() || node->IsCtrlVar()) return false;
D
dzhwinter 已提交
263 264 265

  bool flag = true;
  // op output force generated in cpu, can not be reused.
D
dzhwinter 已提交
266 267
  for (auto* op : node->inputs) {
    if (op->Op()->HasAttr("force_cpu")) {
D
dzhwinter 已提交
268 269
      flag &= framework::AttrReader(op->Op()->GetAttrMap())
                  .Get<bool>("force_cpu") == 0;
D
dzhwinter 已提交
270 271
    }
  }
D
dzhwinter 已提交
272 273
  // var desc validation.
  flag &= NodeCanReused(*node->Var());
D
dzhwinter 已提交
274 275 276 277 278
  return flag;
}

bool NodeCanReused(const VarDesc& node) {
  auto type = node.GetType();
D
dzhwinter 已提交
279 280 281 282 283 284
  if (!(type == proto::VarType::LOD_TENSOR ||
        type == proto::VarType::SELECTED_ROWS ||
        type == proto::VarType::LOD_TENSOR_ARRAY)) {
    return false;
  }
  if (node.Persistable() || node.GetShape().empty()) {
D
dzhwinter 已提交
285 286 287 288 289 290
    return false;
  }
  // vars can be @EMPTY@, @LR_DECAY_REUSE_ID@. For example, while_grad
  std::string name = node.Name();
  if (!name.empty() && name[0] == '@' && name[name.size() - 1] == '@')
    return false;
D
dzhwinter 已提交
291 292 293 294 295 296 297 298 299 300 301 302 303
  return true;
}

bool OpHasSubBlock(OpDesc* desc) {
  const AttributeMap& attrs = desc->GetAttrMap();
  for (auto& attr : attrs) {
    if (attr.second.type() == typeid(BlockDesc*) ||             // NOLINT
        attr.second.type() == typeid(std::vector<BlockDesc*>))  // NOLINT
      return true;
  }
  return false;
}

D
dzhwinter 已提交
304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471
ControlFlowGraph::ControlFlowGraph(const ir::Graph& graph) {
  ops_ = SortOpLikeDescOrder(graph);
  ConnectNodes();
}

void ControlFlowGraph::BuildCFGGraph() {
  // FIXME(dzh): same effect with ConnectNodes, but use the control
  // link to build dependency graph, it goes wrong in transformer.
  for (ir::Node* op : ops_) {
    for (auto& input_var : op->inputs) {
      if (!input_var->inputs.empty()) {
        PADDLE_ENFORCE(
            input_var->inputs.size() == 1 && input_var->inputs[0]->IsOp(),
            "Preceding Op Node of Var Node must be unique");
        auto* pred_op = input_var->inputs[0];
        if (pred_op->Op() != nullptr) {
          predecessors_[op].insert(pred_op);
          successors_[pred_op].insert(op);
        }
      }
      if (input_var->IsVar() && !input_var->IsCtrlVar()) {
        uses_[op].insert(input_var->Name());
      }
    }
    for (auto& output_var : op->outputs) {
      // output var may be used by many op
      for (auto* succ_op : output_var->outputs) {
        if (succ_op->Op() != nullptr) {
          successors_[op].insert(succ_op);
          predecessors_[succ_op].insert(op);
        }
      }
      if (output_var->IsVar() && !output_var->IsCtrlVar()) {
        defs_[op].insert(output_var->Name());
      }
    }
  }
}

void ControlFlowGraph::ConnectNodes() {
  for (size_t i = 0; i < ops_.size(); ++i) {
    auto& op = ops_[i];
    try {
      auto& next_op = ops_.at(i + 1);
      successors_[op].insert(next_op);
      predecessors_[next_op].insert(op);
    } catch (...) {
      // do nothing
    }

    FilterVariables(op->inputs,
                    [&](ir::Node* var) { uses_[op].emplace(var->Name()); });

    FilterVariables(op->outputs,
                    [&](ir::Node* var) { defs_[op].emplace(var->Name()); });
  }
}

void ControlFlowGraph::LiveVariableAnalysis() {
  // NOTE(dzh): variable liveless analysis (a.k.a reversed_ops algorithm)
  // compute the liveness of for each variable though reversed_ops algorithm.
  // It iterates the operators from end to begin, compute the live in/live out
  // variable set for each op, then the diff between in/out will be used for
  // the variable reuse. For detail refer to
  // http://www.cs.cornell.edu/courses/cs4120/2013fa/lectures/lec26-fa13.pdf
  std::list<ir::Node*> work_list(ops_.rbegin(), ops_.rend());
  while (!work_list.empty()) {
    ir::Node* op = work_list.front();
    work_list.pop_front();
    // get the live_in calculated before. Empty if first.
    auto prev_live_in = std::move(live_in_[op]);
    for (auto& s : successors_[op]) {
      for (auto& var : live_in_[s]) {
        live_out_[op].insert(var);
      }
    }
    for (auto& var : uses_[op]) {
      live_in_[op].insert(var);
    }
    for (auto& var : live_out_[op]) {
      live_in_[op].insert(var);
    }
    for (auto& var : defs_[op]) {
      live_in_[op].erase(var);
    }

    // If the live_in is not changed, then the liveness analysis of
    // predecessors is completed.
    //
    // Otherwise, recalculate the predecessors liveness
    if (live_in_[op] != prev_live_in) {
      for (auto& pre : predecessors_[op]) {
        work_list.push_back(pre);
      }
    }
  }
}

void ControlFlowGraph::RenameVarInCFGGraph(const std::string& old_node,
                                           const std::string& new_node,
                                           int begin_idx) {
  // update graph from begin idx to the end
  for (size_t i = begin_idx; i != ops_.size(); ++i) {
    auto* op = ops_[i];
    if (uses_[op].find(old_node) != uses_[op].end()) {
      uses_[op].erase(old_node);
      uses_[op].insert(new_node);
    }
    if (defs_[op].find(old_node) != defs_[op].end()) {
      defs_[op].erase(old_node);
      defs_[op].insert(new_node);
    }
    if (live_in_[op].find(old_node) != live_in_[op].end()) {
      live_in_[op].erase(old_node);
      live_in_[op].insert(new_node);
    }
    if (live_out_[op].find(old_node) != live_out_[op].end()) {
      live_out_[op].erase(old_node);
      live_out_[op].insert(new_node);
    }
  }
}

const std::set<std::string> ControlFlowGraph::LiveIn(ir::Node* op) const {
  auto it = live_in_.find(op);
  PADDLE_ENFORCE(
      it != live_in_.end(),
      string::Sprintf("Expect %s in live_in, but Not Found.", op->Name()));
  return it->second;
}

const std::set<std::string> ControlFlowGraph::LiveOut(ir::Node* op) const {
  auto it = live_out_.find(op);
  PADDLE_ENFORCE(
      it != live_out_.end(),
      string::Sprintf("Expect %s in live_out, but Not Found.", op->Name()));
  return it->second;
}

const std::set<std::string> ControlFlowGraph::Use(ir::Node* op) const {
  auto it = uses_.find(op);
  PADDLE_ENFORCE(
      it != uses_.end(),
      string::Sprintf("Expect %s in live_out, but Not Found.", op->Name()));
  return it->second;
}

const std::vector<ir::Node*> ControlFlowGraph::Ops() const { return ops_; }

std::vector<ir::Node*>& ControlFlowGraph::Ops() { return ops_; }

ir::Node* ControlFlowGraph::GetNodeByName(const std::string& name,
                                          ir::Node* op) const {
  // in ssa-graph, different version nodes have same name,
  // this function get the latest version var before target op
  // It may return nullptr, such as data node.
  ir::Node* found_node = nullptr;
  for (auto* node : ops_) {
    if (node == op) break;
    for (auto& output : node->outputs) {
      if (output->Name() == name) {
        found_node = output;
      }
    }
  }
  return found_node;
}

D
dzhwinter 已提交
472 473 474
}  // namespace details
}  // namespace framework
}  // namespace paddle