Mercurial > hg > graal-jvmci-8
comparison src/share/vm/opto/block.hpp @ 418:72c5366e5d86
6743900: frequency based block layout
Summary: post-register allocation pass that drives block layout by edge frequencies
Reviewed-by: never, kvn
author | rasbold |
---|---|
date | Thu, 06 Nov 2008 14:59:10 -0800 |
parents | a61af66fc99e |
children | ad8c8ca4ab0f |
comparison
equal
deleted
inserted
replaced
417:f4fe12e429a4 | 418:72c5366e5d86 |
---|---|
73 Block *rpop() { Block *b = _blocks[0]; _blocks[0]=_blocks[--_cnt]; return b;} | 73 Block *rpop() { Block *b = _blocks[0]; _blocks[0]=_blocks[--_cnt]; return b;} |
74 void remove( uint i ); | 74 void remove( uint i ); |
75 void insert( uint i, Block *n ); | 75 void insert( uint i, Block *n ); |
76 uint size() const { return _cnt; } | 76 uint size() const { return _cnt; } |
77 void reset() { _cnt = 0; } | 77 void reset() { _cnt = 0; } |
78 void print(); | |
78 }; | 79 }; |
79 | 80 |
80 | 81 |
81 class CFGElement : public ResourceObj { | 82 class CFGElement : public ResourceObj { |
82 public: | 83 public: |
127 | 128 |
128 CFGLoop *_loop; // Loop to which this block belongs | 129 CFGLoop *_loop; // Loop to which this block belongs |
129 uint _rpo; // Number in reverse post order walk | 130 uint _rpo; // Number in reverse post order walk |
130 | 131 |
131 virtual bool is_block() { return true; } | 132 virtual bool is_block() { return true; } |
132 float succ_prob(uint i); // return probability of i'th successor | 133 float succ_prob(uint i); // return probability of i'th successor |
134 int num_fall_throughs(); // How many fall-through candidate this block has | |
135 void update_uncommon_branch(Block* un); // Lower branch prob to uncommon code | |
136 bool succ_fall_through(uint i); // Is successor "i" is a fall-through candidate | |
137 Block* lone_fall_through(); // Return lone fall-through Block or null | |
133 | 138 |
134 Block* dom_lca(Block* that); // Compute LCA in dominator tree. | 139 Block* dom_lca(Block* that); // Compute LCA in dominator tree. |
135 #ifdef ASSERT | 140 #ifdef ASSERT |
136 bool dominates(Block* that) { | 141 bool dominates(Block* that) { |
137 int dom_diff = this->_dom_depth - that->_dom_depth; | 142 int dom_diff = this->_dom_depth - that->_dom_depth; |
142 #endif | 147 #endif |
143 | 148 |
144 // Report the alignment required by this block. Must be a power of 2. | 149 // Report the alignment required by this block. Must be a power of 2. |
145 // The previous block will insert nops to get this alignment. | 150 // The previous block will insert nops to get this alignment. |
146 uint code_alignment(); | 151 uint code_alignment(); |
152 uint compute_loop_alignment(); | |
147 | 153 |
148 // BLOCK_FREQUENCY is a sentinel to mark uses of constant block frequencies. | 154 // BLOCK_FREQUENCY is a sentinel to mark uses of constant block frequencies. |
149 // It is currently also used to scale such frequencies relative to | 155 // It is currently also used to scale such frequencies relative to |
150 // FreqCountInvocations relative to the old value of 1500. | 156 // FreqCountInvocations relative to the old value of 1500. |
151 #define BLOCK_FREQUENCY(f) ((f * (float) 1500) / FreqCountInvocations) | 157 #define BLOCK_FREQUENCY(f) ((f * (float) 1500) / FreqCountInvocations) |
182 if( max_pad > 0 ) { | 188 if( max_pad > 0 ) { |
183 assert(is_power_of_2(max_pad+relocInfo::addr_unit()), ""); | 189 assert(is_power_of_2(max_pad+relocInfo::addr_unit()), ""); |
184 int current_alignment = current_offset & max_pad; | 190 int current_alignment = current_offset & max_pad; |
185 if( current_alignment != 0 ) { | 191 if( current_alignment != 0 ) { |
186 uint padding = (block_alignment-current_alignment) & max_pad; | 192 uint padding = (block_alignment-current_alignment) & max_pad; |
187 if( !head()->is_Loop() || | 193 if( has_loop_alignment() && |
188 padding <= (uint)MaxLoopPad || | 194 padding > (uint)MaxLoopPad && |
189 first_inst_size() > padding ) { | 195 first_inst_size() <= padding ) { |
190 return padding; | 196 return 0; |
191 } | 197 } |
198 return padding; | |
192 } | 199 } |
193 } | 200 } |
194 return 0; | 201 return 0; |
195 } | 202 } |
196 | 203 |
199 // Phis or MergeMems. Such blocks are discovered and marked during the | 206 // Phis or MergeMems. Such blocks are discovered and marked during the |
200 // RemoveEmpty phase, and elided during Output. | 207 // RemoveEmpty phase, and elided during Output. |
201 bool _connector; | 208 bool _connector; |
202 void set_connector() { _connector = true; } | 209 void set_connector() { _connector = true; } |
203 bool is_connector() const { return _connector; }; | 210 bool is_connector() const { return _connector; }; |
211 | |
212 // Loop_alignment will be set for blocks which are at the top of loops. | |
213 // The block layout pass may rotate loops such that the loop head may not | |
214 // be the sequentially first block of the loop encountered in the linear | |
215 // list of blocks. If the layout pass is not run, loop alignment is set | |
216 // for each block which is the head of a loop. | |
217 uint _loop_alignment; | |
218 void set_loop_alignment(Block *loop_top) { | |
219 uint new_alignment = loop_top->compute_loop_alignment(); | |
220 if (new_alignment > _loop_alignment) { | |
221 _loop_alignment = new_alignment; | |
222 } | |
223 } | |
224 uint loop_alignment() const { return _loop_alignment; } | |
225 bool has_loop_alignment() const { return loop_alignment() > 0; } | |
204 | 226 |
205 // Create a new Block with given head Node. | 227 // Create a new Block with given head Node. |
206 // Creates the (empty) predecessor arrays. | 228 // Creates the (empty) predecessor arrays. |
207 Block( Arena *a, Node *headnode ) | 229 Block( Arena *a, Node *headnode ) |
208 : CFGElement(), | 230 : CFGElement(), |
217 _freg_pressure(0), | 239 _freg_pressure(0), |
218 _fhrp_index(1), | 240 _fhrp_index(1), |
219 _raise_LCA_mark(0), | 241 _raise_LCA_mark(0), |
220 _raise_LCA_visited(0), | 242 _raise_LCA_visited(0), |
221 _first_inst_size(999999), | 243 _first_inst_size(999999), |
222 _connector(false) { | 244 _connector(false), |
245 _loop_alignment(0) { | |
223 _nodes.push(headnode); | 246 _nodes.push(headnode); |
224 } | 247 } |
225 | 248 |
226 // Index of 'end' Node | 249 // Index of 'end' Node |
227 uint end_idx() const { | 250 uint end_idx() const { |
273 s = s->_succs[0]; | 296 s = s->_succs[0]; |
274 } | 297 } |
275 return s; | 298 return s; |
276 } | 299 } |
277 | 300 |
301 // Return true if b is a successor of this block | |
302 bool has_successor(Block* b) const { | |
303 for (uint i = 0; i < _num_succs; i++ ) { | |
304 if (non_connector_successor(i) == b) { | |
305 return true; | |
306 } | |
307 } | |
308 return false; | |
309 } | |
310 | |
278 // Successor block, after forwarding through connectors | 311 // Successor block, after forwarding through connectors |
279 Block* non_connector_successor(int i) const { | 312 Block* non_connector_successor(int i) const { |
280 return _succs[i]->non_connector(); | 313 return _succs[i]->non_connector(); |
281 } | 314 } |
282 | 315 |
317 // Set the basic block for pinned Nodes | 350 // Set the basic block for pinned Nodes |
318 void schedule_pinned_nodes( VectorSet &visited ); | 351 void schedule_pinned_nodes( VectorSet &visited ); |
319 | 352 |
320 // I'll need a few machine-specific GotoNodes. Clone from this one. | 353 // I'll need a few machine-specific GotoNodes. Clone from this one. |
321 MachNode *_goto; | 354 MachNode *_goto; |
322 void insert_goto_at(uint block_no, uint succ_no); | |
323 | 355 |
324 Block* insert_anti_dependences(Block* LCA, Node* load, bool verify = false); | 356 Block* insert_anti_dependences(Block* LCA, Node* load, bool verify = false); |
325 void verify_anti_dependences(Block* LCA, Node* load) { | 357 void verify_anti_dependences(Block* LCA, Node* load) { |
326 assert(LCA == _bbs[load->_idx], "should already be scheduled"); | 358 assert(LCA == _bbs[load->_idx], "should already be scheduled"); |
327 insert_anti_dependences(LCA, load, true); | 359 insert_anti_dependences(LCA, load, true); |
377 Block* hoist_to_cheaper_block(Block* LCA, Block* early, Node* self); | 409 Block* hoist_to_cheaper_block(Block* LCA, Block* early, Node* self); |
378 | 410 |
379 // Compute the instruction global latency with a backwards walk | 411 // Compute the instruction global latency with a backwards walk |
380 void ComputeLatenciesBackwards(VectorSet &visited, Node_List &stack); | 412 void ComputeLatenciesBackwards(VectorSet &visited, Node_List &stack); |
381 | 413 |
414 // Set loop alignment | |
415 void set_loop_alignment(); | |
416 | |
382 // Remove empty basic blocks | 417 // Remove empty basic blocks |
383 void RemoveEmpty(); | 418 void remove_empty(); |
384 bool MoveToNext(Block* bx, uint b_index); | 419 void fixup_flow(); |
385 void MoveToEnd(Block* bx, uint b_index); | 420 bool move_to_next(Block* bx, uint b_index); |
421 void move_to_end(Block* bx, uint b_index); | |
422 void insert_goto_at(uint block_no, uint succ_no); | |
386 | 423 |
387 // Check for NeverBranch at block end. This needs to become a GOTO to the | 424 // Check for NeverBranch at block end. This needs to become a GOTO to the |
388 // true target. NeverBranch are treated as a conditional branch that always | 425 // true target. NeverBranch are treated as a conditional branch that always |
389 // goes the same direction for most of the optimizer and are used to give a | 426 // goes the same direction for most of the optimizer and are used to give a |
390 // fake exit path to infinite loops. At this late stage they need to turn | 427 // fake exit path to infinite loops. At this late stage they need to turn |
411 bool trace_opto_pipelining() const { return false; } | 448 bool trace_opto_pipelining() const { return false; } |
412 #endif | 449 #endif |
413 }; | 450 }; |
414 | 451 |
415 | 452 |
416 //------------------------------UnionFindInfo---------------------------------- | 453 //------------------------------UnionFind-------------------------------------- |
417 // Map Block indices to a block-index for a cfg-cover. | 454 // Map Block indices to a block-index for a cfg-cover. |
418 // Array lookup in the optimized case. | 455 // Array lookup in the optimized case. |
419 class UnionFind : public ResourceObj { | 456 class UnionFind : public ResourceObj { |
420 uint _cnt, _max; | 457 uint _cnt, _max; |
421 uint* _indices; | 458 uint* _indices; |
506 #ifndef PRODUCT | 543 #ifndef PRODUCT |
507 void dump( ) const; | 544 void dump( ) const; |
508 void dump_tree() const; | 545 void dump_tree() const; |
509 #endif | 546 #endif |
510 }; | 547 }; |
548 | |
549 | |
550 //----------------------------------CFGEdge------------------------------------ | |
551 // A edge between two basic blocks that will be embodied by a branch or a | |
552 // fall-through. | |
553 class CFGEdge : public ResourceObj { | |
554 private: | |
555 Block * _from; // Source basic block | |
556 Block * _to; // Destination basic block | |
557 float _freq; // Execution frequency (estimate) | |
558 int _state; | |
559 bool _infrequent; | |
560 int _from_pct; | |
561 int _to_pct; | |
562 | |
563 // Private accessors | |
564 int from_pct() const { return _from_pct; } | |
565 int to_pct() const { return _to_pct; } | |
566 int from_infrequent() const { return from_pct() < BlockLayoutMinDiamondPercentage; } | |
567 int to_infrequent() const { return to_pct() < BlockLayoutMinDiamondPercentage; } | |
568 | |
569 public: | |
570 enum { | |
571 open, // initial edge state; unprocessed | |
572 connected, // edge used to connect two traces together | |
573 interior // edge is interior to trace (could be backedge) | |
574 }; | |
575 | |
576 CFGEdge(Block *from, Block *to, float freq, int from_pct, int to_pct) : | |
577 _from(from), _to(to), _freq(freq), | |
578 _from_pct(from_pct), _to_pct(to_pct), _state(open) { | |
579 _infrequent = from_infrequent() || to_infrequent(); | |
580 } | |
581 | |
582 float freq() const { return _freq; } | |
583 Block* from() const { return _from; } | |
584 Block* to () const { return _to; } | |
585 int infrequent() const { return _infrequent; } | |
586 int state() const { return _state; } | |
587 | |
588 void set_state(int state) { _state = state; } | |
589 | |
590 #ifndef PRODUCT | |
591 void dump( ) const; | |
592 #endif | |
593 }; | |
594 | |
595 | |
596 //-----------------------------------Trace------------------------------------- | |
597 // An ordered list of basic blocks. | |
598 class Trace : public ResourceObj { | |
599 private: | |
600 uint _id; // Unique Trace id (derived from initial block) | |
601 Block ** _next_list; // Array mapping index to next block | |
602 Block ** _prev_list; // Array mapping index to previous block | |
603 Block * _first; // First block in the trace | |
604 Block * _last; // Last block in the trace | |
605 | |
606 // Return the block that follows "b" in the trace. | |
607 Block * next(Block *b) const { return _next_list[b->_pre_order]; } | |
608 void set_next(Block *b, Block *n) const { _next_list[b->_pre_order] = n; } | |
609 | |
610 // Return the block that preceeds "b" in the trace. | |
611 Block * prev(Block *b) const { return _prev_list[b->_pre_order]; } | |
612 void set_prev(Block *b, Block *p) const { _prev_list[b->_pre_order] = p; } | |
613 | |
614 // We've discovered a loop in this trace. Reset last to be "b", and first as | |
615 // the block following "b | |
616 void break_loop_after(Block *b) { | |
617 _last = b; | |
618 _first = next(b); | |
619 set_prev(_first, NULL); | |
620 set_next(_last, NULL); | |
621 } | |
622 | |
623 public: | |
624 | |
625 Trace(Block *b, Block **next_list, Block **prev_list) : | |
626 _first(b), | |
627 _last(b), | |
628 _next_list(next_list), | |
629 _prev_list(prev_list), | |
630 _id(b->_pre_order) { | |
631 set_next(b, NULL); | |
632 set_prev(b, NULL); | |
633 }; | |
634 | |
635 // Return the id number | |
636 uint id() const { return _id; } | |
637 void set_id(uint id) { _id = id; } | |
638 | |
639 // Return the first block in the trace | |
640 Block * first_block() const { return _first; } | |
641 | |
642 // Return the last block in the trace | |
643 Block * last_block() const { return _last; } | |
644 | |
645 // Insert a trace in the middle of this one after b | |
646 void insert_after(Block *b, Trace *tr) { | |
647 set_next(tr->last_block(), next(b)); | |
648 if (next(b) != NULL) { | |
649 set_prev(next(b), tr->last_block()); | |
650 } | |
651 | |
652 set_next(b, tr->first_block()); | |
653 set_prev(tr->first_block(), b); | |
654 | |
655 if (b == _last) { | |
656 _last = tr->last_block(); | |
657 } | |
658 } | |
659 | |
660 void insert_before(Block *b, Trace *tr) { | |
661 Block *p = prev(b); | |
662 assert(p != NULL, "use append instead"); | |
663 insert_after(p, tr); | |
664 } | |
665 | |
666 // Append another trace to this one. | |
667 void append(Trace *tr) { | |
668 insert_after(_last, tr); | |
669 } | |
670 | |
671 // Append a block at the end of this trace | |
672 void append(Block *b) { | |
673 set_next(_last, b); | |
674 set_prev(b, _last); | |
675 _last = b; | |
676 } | |
677 | |
678 // Adjust the the blocks in this trace | |
679 void fixup_blocks(PhaseCFG &cfg); | |
680 bool backedge(CFGEdge *e); | |
681 | |
682 #ifndef PRODUCT | |
683 void dump( ) const; | |
684 #endif | |
685 }; | |
686 | |
687 //------------------------------PhaseBlockLayout------------------------------- | |
688 // Rearrange blocks into some canonical order, based on edges and their frequencies | |
689 class PhaseBlockLayout : public Phase { | |
690 PhaseCFG &_cfg; // Control flow graph | |
691 | |
692 GrowableArray<CFGEdge *> *edges; | |
693 Trace **traces; | |
694 Block **next; | |
695 Block **prev; | |
696 UnionFind *uf; | |
697 | |
698 // Given a block, find its encompassing Trace | |
699 Trace * trace(Block *b) { | |
700 return traces[uf->Find_compress(b->_pre_order)]; | |
701 } | |
702 public: | |
703 PhaseBlockLayout(PhaseCFG &cfg); | |
704 | |
705 void find_edges(); | |
706 void grow_traces(); | |
707 void merge_traces(bool loose_connections); | |
708 void reorder_traces(int count); | |
709 void union_traces(Trace* from, Trace* to); | |
710 }; |