1 /*
2  * Copyright (C) 2016 The Android Open Source Project
3  *
4  * Licensed under the Apache License, Version 2.0 (the "License");
5  * you may not use this file except in compliance with the License.
6  * You may obtain a copy of the License at
7  *
8  *      http://www.apache.org/licenses/LICENSE-2.0
9  *
10  * Unless required by applicable law or agreed to in writing, software
11  * distributed under the License is distributed on an "AS IS" BASIS,
12  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13  * See the License for the specific language governing permissions and
14  * limitations under the License.
15  */
16 
17 #ifndef ART_COMPILER_OPTIMIZING_SCHEDULER_H_
18 #define ART_COMPILER_OPTIMIZING_SCHEDULER_H_
19 
20 #include <fstream>
21 
22 #include "base/scoped_arena_allocator.h"
23 #include "base/scoped_arena_containers.h"
24 #include "base/time_utils.h"
25 #include "code_generator.h"
26 #include "driver/compiler_driver.h"
27 #include "load_store_analysis.h"
28 #include "nodes.h"
29 #include "optimization.h"
30 
31 namespace art {
32 
33 // General description of instruction scheduling.
34 //
35 // This pass tries to improve the quality of the generated code by reordering
36 // instructions in the graph to avoid execution delays caused by execution
37 // dependencies.
38 // Currently, scheduling is performed at the block level, so no `HInstruction`
39 // ever leaves its block in this pass.
40 //
41 // The scheduling process iterates through blocks in the graph. For blocks that
42 // we can and want to schedule:
43 // 1) Build a dependency graph for instructions.
44 //    It includes data dependencies (inputs/uses), but also environment
45 //    dependencies and side-effect dependencies.
46 // 2) Schedule the dependency graph.
47 //    This is a topological sort of the dependency graph, using heuristics to
48 //    decide what node to scheduler first when there are multiple candidates.
49 //
50 // A few factors impacting the quality of the scheduling are:
51 // - The heuristics used to decide what node to schedule in the topological sort
52 //   when there are multiple valid candidates. There is a wide range of
53 //   complexity possible here, going from a simple model only considering
54 //   latencies, to a super detailed CPU pipeline model.
55 // - Fewer dependencies in the dependency graph give more freedom for the
56 //   scheduling heuristics. For example de-aliasing can allow possibilities for
57 //   reordering of memory accesses.
58 // - The level of abstraction of the IR. It is easier to evaluate scheduling for
59 //   IRs that translate to a single assembly instruction than for IRs
60 //   that generate multiple assembly instructions or generate different code
61 //   depending on properties of the IR.
62 // - Scheduling is performed before register allocation, it is not aware of the
63 //   impact of moving instructions on register allocation.
64 //
65 //
66 // The scheduling code uses the terms predecessors, successors, and dependencies.
67 // This can be confusing at times, so here are clarifications.
68 // These terms are used from the point of view of the program dependency graph. So
69 // the inputs of an instruction are part of its dependencies, and hence part its
70 // predecessors. So the uses of an instruction are (part of) its successors.
71 // (Side-effect dependencies can yield predecessors or successors that are not
72 // inputs or uses.)
73 //
74 // Here is a trivial example. For the Java code:
75 //
76 //    int a = 1 + 2;
77 //
78 // we would have the instructions
79 //
80 //    i1 HIntConstant 1
81 //    i2 HIntConstant 2
82 //    i3 HAdd [i1,i2]
83 //
84 // `i1` and `i2` are predecessors of `i3`.
85 // `i3` is a successor of `i1` and a successor of `i2`.
86 // In a scheduling graph for this code we would have three nodes `n1`, `n2`,
87 // and `n3` (respectively for instructions `i1`, `i1`, and `i3`).
88 // Conceptually the program dependency graph for this would contain two edges
89 //
90 //    n1 -> n3
91 //    n2 -> n3
92 //
93 // Since we schedule backwards (starting from the last instruction in each basic
94 // block), the implementation of nodes keeps a list of pointers their
95 // predecessors. So `n3` would keep pointers to its predecessors `n1` and `n2`.
96 //
97 // Node dependencies are also referred to from the program dependency graph
98 // point of view: we say that node `B` immediately depends on `A` if there is an
99 // edge from `A` to `B` in the program dependency graph. `A` is a predecessor of
100 // `B`, `B` is a successor of `A`. In the example above `n3` depends on `n1` and
101 // `n2`.
102 // Since nodes in the scheduling graph keep a list of their predecessors, node
103 // `B` will have a pointer to its predecessor `A`.
104 // As we schedule backwards, `B` will be selected for scheduling before `A` is.
105 //
106 // So the scheduling for the example above could happen as follow
107 //
108 //    |---------------------------+------------------------|
109 //    | candidates for scheduling | instructions scheduled |
110 //    | --------------------------+------------------------|
111 //
112 // The only node without successors is `n3`, so it is the only initial
113 // candidate.
114 //
115 //    | n3                        | (none)                 |
116 //
117 // We schedule `n3` as the last (and only) instruction. All its predecessors
118 // that do not have any unscheduled successors become candidate. That is, `n1`
119 // and `n2` become candidates.
120 //
121 //    | n1, n2                    | n3                     |
122 //
123 // One of the candidates is selected. In practice this is where scheduling
124 // heuristics kick in, to decide which of the candidates should be selected.
125 // In this example, let it be `n1`. It is scheduled before previously scheduled
126 // nodes (in program order). There are no other nodes to add to the list of
127 // candidates.
128 //
129 //    | n2                        | n1                     |
130 //    |                           | n3                     |
131 //
132 // The only candidate available for scheduling is `n2`. Schedule it before
133 // (in program order) the previously scheduled nodes.
134 //
135 //    | (none)                    | n2                     |
136 //    |                           | n1                     |
137 //    |                           | n3                     |
138 //    |---------------------------+------------------------|
139 //
140 // So finally the instructions will be executed in the order `i2`, `i1`, and `i3`.
141 // In this trivial example, it does not matter which of `i1` and `i2` is
142 // scheduled first since they are constants. However the same process would
143 // apply if `i1` and `i2` were actual operations (for example `HMul` and `HDiv`).
144 
145 // Set to true to have instruction scheduling dump scheduling graphs to the file
146 // `scheduling_graphs.dot`. See `SchedulingGraph::DumpAsDotGraph()`.
147 static constexpr bool kDumpDotSchedulingGraphs = false;
148 
149 // Typically used as a default instruction latency.
150 static constexpr uint32_t kGenericInstructionLatency = 1;
151 
152 class HScheduler;
153 
154 /**
155  * A node representing an `HInstruction` in the `SchedulingGraph`.
156  */
157 class SchedulingNode : public DeletableArenaObject<kArenaAllocScheduler> {
158  public:
SchedulingNode(HInstruction * instr,ScopedArenaAllocator * allocator,bool is_scheduling_barrier)159   SchedulingNode(HInstruction* instr, ScopedArenaAllocator* allocator, bool is_scheduling_barrier)
160       : latency_(0),
161         internal_latency_(0),
162         critical_path_(0),
163         instruction_(instr),
164         is_scheduling_barrier_(is_scheduling_barrier),
165         data_predecessors_(allocator->Adapter(kArenaAllocScheduler)),
166         other_predecessors_(allocator->Adapter(kArenaAllocScheduler)),
167         num_unscheduled_successors_(0) {
168     data_predecessors_.reserve(kPreallocatedPredecessors);
169   }
170 
AddDataPredecessor(SchedulingNode * predecessor)171   void AddDataPredecessor(SchedulingNode* predecessor) {
172     data_predecessors_.push_back(predecessor);
173     predecessor->num_unscheduled_successors_++;
174   }
175 
GetDataPredecessors()176   const ScopedArenaVector<SchedulingNode*>& GetDataPredecessors() const {
177     return data_predecessors_;
178   }
179 
AddOtherPredecessor(SchedulingNode * predecessor)180   void AddOtherPredecessor(SchedulingNode* predecessor) {
181     other_predecessors_.push_back(predecessor);
182     predecessor->num_unscheduled_successors_++;
183   }
184 
GetOtherPredecessors()185   const ScopedArenaVector<SchedulingNode*>& GetOtherPredecessors() const {
186     return other_predecessors_;
187   }
188 
DecrementNumberOfUnscheduledSuccessors()189   void DecrementNumberOfUnscheduledSuccessors() {
190     num_unscheduled_successors_--;
191   }
192 
MaybeUpdateCriticalPath(uint32_t other_critical_path)193   void MaybeUpdateCriticalPath(uint32_t other_critical_path) {
194     critical_path_ = std::max(critical_path_, other_critical_path);
195   }
196 
HasUnscheduledSuccessors()197   bool HasUnscheduledSuccessors() const {
198     return num_unscheduled_successors_ != 0;
199   }
200 
GetInstruction()201   HInstruction* GetInstruction() const { return instruction_; }
GetLatency()202   uint32_t GetLatency() const { return latency_; }
SetLatency(uint32_t latency)203   void SetLatency(uint32_t latency) { latency_ = latency; }
GetInternalLatency()204   uint32_t GetInternalLatency() const { return internal_latency_; }
SetInternalLatency(uint32_t internal_latency)205   void SetInternalLatency(uint32_t internal_latency) { internal_latency_ = internal_latency; }
GetCriticalPath()206   uint32_t GetCriticalPath() const { return critical_path_; }
IsSchedulingBarrier()207   bool IsSchedulingBarrier() const { return is_scheduling_barrier_; }
208 
209  private:
210   // The latency of this node. It represents the latency between the moment the
211   // last instruction for this node has executed to the moment the result
212   // produced by this node is available to users.
213   uint32_t latency_;
214   // This represents the time spent *within* the generated code for this node.
215   // It should be zero for nodes that only generate a single instruction.
216   uint32_t internal_latency_;
217 
218   // The critical path from this instruction to the end of scheduling. It is
219   // used by the scheduling heuristics to measure the priority of this instruction.
220   // It is defined as
221   //     critical_path_ = latency_ + max((use.internal_latency_ + use.critical_path_) for all uses)
222   // (Note that here 'uses' is equivalent to 'data successors'. Also see comments in
223   // `HScheduler::Schedule(SchedulingNode* scheduling_node)`).
224   uint32_t critical_path_;
225 
226   // The instruction that this node represents.
227   HInstruction* const instruction_;
228 
229   // If a node is scheduling barrier, other nodes cannot be scheduled before it.
230   const bool is_scheduling_barrier_;
231 
232   // The lists of predecessors. They cannot be scheduled before this node. Once
233   // this node is scheduled, we check whether any of its predecessors has become a
234   // valid candidate for scheduling.
235   // Predecessors in `data_predecessors_` are data dependencies. Those in
236   // `other_predecessors_` contain side-effect dependencies, environment
237   // dependencies, and scheduling barrier dependencies.
238   ScopedArenaVector<SchedulingNode*> data_predecessors_;
239   ScopedArenaVector<SchedulingNode*> other_predecessors_;
240 
241   // The number of unscheduled successors for this node. This number is
242   // decremented as successors are scheduled. When it reaches zero this node
243   // becomes a valid candidate to schedule.
244   uint32_t num_unscheduled_successors_;
245 
246   static constexpr size_t kPreallocatedPredecessors = 4;
247 };
248 
249 /*
250  * Directed acyclic graph for scheduling.
251  */
252 class SchedulingGraph : public ValueObject {
253  public:
SchedulingGraph(const HScheduler * scheduler,ScopedArenaAllocator * allocator)254   SchedulingGraph(const HScheduler* scheduler, ScopedArenaAllocator* allocator)
255       : scheduler_(scheduler),
256         allocator_(allocator),
257         contains_scheduling_barrier_(false),
258         nodes_map_(allocator_->Adapter(kArenaAllocScheduler)),
259         heap_location_collector_(nullptr) {}
260 
261   SchedulingNode* AddNode(HInstruction* instr, bool is_scheduling_barrier = false) {
262     std::unique_ptr<SchedulingNode> node(
263         new (allocator_) SchedulingNode(instr, allocator_, is_scheduling_barrier));
264     SchedulingNode* result = node.get();
265     nodes_map_.Insert(std::make_pair(instr, std::move(node)));
266     contains_scheduling_barrier_ |= is_scheduling_barrier;
267     AddDependencies(instr, is_scheduling_barrier);
268     return result;
269   }
270 
Clear()271   void Clear() {
272     nodes_map_.Clear();
273     contains_scheduling_barrier_ = false;
274   }
275 
SetHeapLocationCollector(const HeapLocationCollector & heap_location_collector)276   void SetHeapLocationCollector(const HeapLocationCollector& heap_location_collector) {
277     heap_location_collector_ = &heap_location_collector;
278   }
279 
GetNode(const HInstruction * instr)280   SchedulingNode* GetNode(const HInstruction* instr) const {
281     auto it = nodes_map_.Find(instr);
282     if (it == nodes_map_.end()) {
283       return nullptr;
284     } else {
285       return it->second.get();
286     }
287   }
288 
289   bool IsSchedulingBarrier(const HInstruction* instruction) const;
290 
291   bool HasImmediateDataDependency(const SchedulingNode* node, const SchedulingNode* other) const;
292   bool HasImmediateDataDependency(const HInstruction* node, const HInstruction* other) const;
293   bool HasImmediateOtherDependency(const SchedulingNode* node, const SchedulingNode* other) const;
294   bool HasImmediateOtherDependency(const HInstruction* node, const HInstruction* other) const;
295 
Size()296   size_t Size() const {
297     return nodes_map_.Size();
298   }
299 
300   // Dump the scheduling graph, in dot file format, appending it to the file
301   // `scheduling_graphs.dot`.
302   void DumpAsDotGraph(const std::string& description,
303                       const ScopedArenaVector<SchedulingNode*>& initial_candidates);
304 
305  protected:
306   void AddDependency(SchedulingNode* node, SchedulingNode* dependency, bool is_data_dependency);
AddDataDependency(SchedulingNode * node,SchedulingNode * dependency)307   void AddDataDependency(SchedulingNode* node, SchedulingNode* dependency) {
308     AddDependency(node, dependency, /*is_data_dependency*/true);
309   }
AddOtherDependency(SchedulingNode * node,SchedulingNode * dependency)310   void AddOtherDependency(SchedulingNode* node, SchedulingNode* dependency) {
311     AddDependency(node, dependency, /*is_data_dependency*/false);
312   }
313   bool HasMemoryDependency(const HInstruction* node, const HInstruction* other) const;
314   bool HasExceptionDependency(const HInstruction* node, const HInstruction* other) const;
315   bool HasSideEffectDependency(const HInstruction* node, const HInstruction* other) const;
316   bool ArrayAccessMayAlias(const HInstruction* node, const HInstruction* other) const;
317   bool FieldAccessMayAlias(const HInstruction* node, const HInstruction* other) const;
318   size_t ArrayAccessHeapLocation(HInstruction* array, HInstruction* index) const;
319   size_t FieldAccessHeapLocation(HInstruction* obj, const FieldInfo* field) const;
320 
321   // Add dependencies nodes for the given `HInstruction`: inputs, environments, and side-effects.
322   void AddDependencies(HInstruction* instruction, bool is_scheduling_barrier = false);
323 
324   const HScheduler* const scheduler_;
325 
326   ScopedArenaAllocator* const allocator_;
327 
328   bool contains_scheduling_barrier_;
329 
330   ScopedArenaHashMap<const HInstruction*, std::unique_ptr<SchedulingNode>> nodes_map_;
331 
332   const HeapLocationCollector* heap_location_collector_;
333 };
334 
335 /*
336  * The visitors derived from this base class are used by schedulers to evaluate
337  * the latencies of `HInstruction`s.
338  */
339 class SchedulingLatencyVisitor : public HGraphDelegateVisitor {
340  public:
341   // This class and its sub-classes will never be used to drive a visit of an
342   // `HGraph` but only to visit `HInstructions` one at a time, so we do not need
343   // to pass a valid graph to `HGraphDelegateVisitor()`.
SchedulingLatencyVisitor()344   SchedulingLatencyVisitor()
345       : HGraphDelegateVisitor(nullptr),
346         last_visited_latency_(0),
347         last_visited_internal_latency_(0) {}
348 
VisitInstruction(HInstruction * instruction)349   void VisitInstruction(HInstruction* instruction) OVERRIDE {
350     LOG(FATAL) << "Error visiting " << instruction->DebugName() << ". "
351         "Architecture-specific scheduling latency visitors must handle all instructions"
352         " (potentially by overriding the generic `VisitInstruction()`.";
353     UNREACHABLE();
354   }
355 
Visit(HInstruction * instruction)356   void Visit(HInstruction* instruction) {
357     instruction->Accept(this);
358   }
359 
CalculateLatency(SchedulingNode * node)360   void CalculateLatency(SchedulingNode* node) {
361     // By default nodes have no internal latency.
362     last_visited_internal_latency_ = 0;
363     Visit(node->GetInstruction());
364   }
365 
GetLastVisitedLatency()366   uint32_t GetLastVisitedLatency() const { return last_visited_latency_; }
GetLastVisitedInternalLatency()367   uint32_t GetLastVisitedInternalLatency() const { return last_visited_internal_latency_; }
368 
369  protected:
370   // The latency of the most recent visited SchedulingNode.
371   // This is for reporting the latency value to the user of this visitor.
372   uint32_t last_visited_latency_;
373   // This represents the time spent *within* the generated code for the most recent visited
374   // SchedulingNode. This is for reporting the internal latency value to the user of this visitor.
375   uint32_t last_visited_internal_latency_;
376 };
377 
378 class SchedulingNodeSelector : public ArenaObject<kArenaAllocScheduler> {
379  public:
380   virtual SchedulingNode* PopHighestPriorityNode(ScopedArenaVector<SchedulingNode*>* nodes,
381                                                  const SchedulingGraph& graph) = 0;
~SchedulingNodeSelector()382   virtual ~SchedulingNodeSelector() {}
383  protected:
DeleteNodeAtIndex(ScopedArenaVector<SchedulingNode * > * nodes,size_t index)384   static void DeleteNodeAtIndex(ScopedArenaVector<SchedulingNode*>* nodes, size_t index) {
385     (*nodes)[index] = nodes->back();
386     nodes->pop_back();
387   }
388 };
389 
390 /*
391  * Select a `SchedulingNode` at random within the candidates.
392  */
393 class RandomSchedulingNodeSelector : public SchedulingNodeSelector {
394  public:
RandomSchedulingNodeSelector()395   RandomSchedulingNodeSelector() : seed_(0) {
396     seed_  = static_cast<uint32_t>(NanoTime());
397     srand(seed_);
398   }
399 
PopHighestPriorityNode(ScopedArenaVector<SchedulingNode * > * nodes,const SchedulingGraph & graph)400   SchedulingNode* PopHighestPriorityNode(ScopedArenaVector<SchedulingNode*>* nodes,
401                                          const SchedulingGraph& graph) OVERRIDE {
402     UNUSED(graph);
403     DCHECK(!nodes->empty());
404     size_t select = rand_r(&seed_) % nodes->size();
405     SchedulingNode* select_node = (*nodes)[select];
406     DeleteNodeAtIndex(nodes, select);
407     return select_node;
408   }
409 
410   uint32_t seed_;
411 };
412 
413 /*
414  * Select a `SchedulingNode` according to critical path information,
415  * with heuristics to favor certain instruction patterns like materialized condition.
416  */
417 class CriticalPathSchedulingNodeSelector : public SchedulingNodeSelector {
418  public:
CriticalPathSchedulingNodeSelector()419   CriticalPathSchedulingNodeSelector() : prev_select_(nullptr) {}
420 
421   SchedulingNode* PopHighestPriorityNode(ScopedArenaVector<SchedulingNode*>* nodes,
422                                          const SchedulingGraph& graph) OVERRIDE;
423 
424  protected:
425   SchedulingNode* GetHigherPrioritySchedulingNode(SchedulingNode* candidate,
426                                                   SchedulingNode* check) const;
427 
428   SchedulingNode* SelectMaterializedCondition(ScopedArenaVector<SchedulingNode*>* nodes,
429                                               const SchedulingGraph& graph) const;
430 
431  private:
432   const SchedulingNode* prev_select_;
433 };
434 
435 class HScheduler {
436  public:
HScheduler(ScopedArenaAllocator * allocator,SchedulingLatencyVisitor * latency_visitor,SchedulingNodeSelector * selector)437   HScheduler(ScopedArenaAllocator* allocator,
438              SchedulingLatencyVisitor* latency_visitor,
439              SchedulingNodeSelector* selector)
440       : allocator_(allocator),
441         latency_visitor_(latency_visitor),
442         selector_(selector),
443         only_optimize_loop_blocks_(true),
444         scheduling_graph_(this, allocator),
445         cursor_(nullptr),
446         candidates_(allocator_->Adapter(kArenaAllocScheduler)) {}
~HScheduler()447   virtual ~HScheduler() {}
448 
449   void Schedule(HGraph* graph);
450 
SetOnlyOptimizeLoopBlocks(bool loop_only)451   void SetOnlyOptimizeLoopBlocks(bool loop_only) { only_optimize_loop_blocks_ = loop_only; }
452 
453   // Instructions can not be rescheduled across a scheduling barrier.
454   virtual bool IsSchedulingBarrier(const HInstruction* instruction) const;
455 
456  protected:
457   void Schedule(HBasicBlock* block);
458   void Schedule(SchedulingNode* scheduling_node);
459   void Schedule(HInstruction* instruction);
460 
461   // Any instruction returning `false` via this method will prevent its
462   // containing basic block from being scheduled.
463   // This method is used to restrict scheduling to instructions that we know are
464   // safe to handle.
465   //
466   // For newly introduced instructions by default HScheduler::IsSchedulable returns false.
467   // HScheduler${ARCH}::IsSchedulable can be overridden to return true for an instruction (see
468   // scheduler_arm64.h for example) if it is safe to schedule it; in this case one *must* also
469   // look at/update HScheduler${ARCH}::IsSchedulingBarrier for this instruction.
470   virtual bool IsSchedulable(const HInstruction* instruction) const;
471   bool IsSchedulable(const HBasicBlock* block) const;
472 
CalculateLatency(SchedulingNode * node)473   void CalculateLatency(SchedulingNode* node) {
474     latency_visitor_->CalculateLatency(node);
475     node->SetLatency(latency_visitor_->GetLastVisitedLatency());
476     node->SetInternalLatency(latency_visitor_->GetLastVisitedInternalLatency());
477   }
478 
479   ScopedArenaAllocator* const allocator_;
480   SchedulingLatencyVisitor* const latency_visitor_;
481   SchedulingNodeSelector* const selector_;
482   bool only_optimize_loop_blocks_;
483 
484   // We instantiate the members below as part of this class to avoid
485   // instantiating them locally for every chunk scheduled.
486   SchedulingGraph scheduling_graph_;
487   // A pointer indicating where the next instruction to be scheduled will be inserted.
488   HInstruction* cursor_;
489   // The list of candidates for scheduling. A node becomes a candidate when all
490   // its predecessors have been scheduled.
491   ScopedArenaVector<SchedulingNode*> candidates_;
492 
493  private:
494   DISALLOW_COPY_AND_ASSIGN(HScheduler);
495 };
496 
IsSchedulingBarrier(const HInstruction * instruction)497 inline bool SchedulingGraph::IsSchedulingBarrier(const HInstruction* instruction) const {
498   return scheduler_->IsSchedulingBarrier(instruction);
499 }
500 
501 class HInstructionScheduling : public HOptimization {
502  public:
503   HInstructionScheduling(HGraph* graph,
504                          InstructionSet instruction_set,
505                          CodeGenerator* cg = nullptr,
506                          const char* name = kInstructionSchedulingPassName)
HOptimization(graph,name)507       : HOptimization(graph, name),
508         codegen_(cg),
509         instruction_set_(instruction_set) {}
510 
Run()511   void Run() {
512     Run(/*only_optimize_loop_blocks*/ true, /*schedule_randomly*/ false);
513   }
514   void Run(bool only_optimize_loop_blocks, bool schedule_randomly);
515 
516   static constexpr const char* kInstructionSchedulingPassName = "scheduler";
517 
518  private:
519   CodeGenerator* const codegen_;
520   const InstructionSet instruction_set_;
521   DISALLOW_COPY_AND_ASSIGN(HInstructionScheduling);
522 };
523 
524 }  // namespace art
525 
526 #endif  // ART_COMPILER_OPTIMIZING_SCHEDULER_H_
527