1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
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
118
119
120
121
122
123
124
125
126
127
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
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
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
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
|
// Copyright 2014 the V8 project authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.
#include "src/compiler/backend/jump-threading.h"
#include "src/compiler/backend/code-generator-impl.h"
namespace v8 {
namespace internal {
namespace compiler {
#define TRACE(...) \
do { \
if (v8_flags.trace_turbo_jt) PrintF(__VA_ARGS__); \
} while (false)
namespace {
struct JumpThreadingState {
bool forwarded;
ZoneVector<RpoNumber>& result;
ZoneStack<RpoNumber>& stack;
void Clear(size_t count) { result.assign(count, unvisited()); }
void PushIfUnvisited(RpoNumber num) {
if (result[num.ToInt()] == unvisited()) {
stack.push(num);
result[num.ToInt()] = onstack();
}
}
void Forward(RpoNumber to) {
RpoNumber from = stack.top();
RpoNumber to_to = result[to.ToInt()];
bool pop = true;
if (to == from) {
TRACE(" xx %d\n", from.ToInt());
result[from.ToInt()] = from;
} else if (to_to == unvisited()) {
TRACE(" fw %d -> %d (recurse)\n", from.ToInt(), to.ToInt());
stack.push(to);
result[to.ToInt()] = onstack();
pop = false; // recurse.
} else if (to_to == onstack()) {
TRACE(" fw %d -> %d (cycle)\n", from.ToInt(), to.ToInt());
result[from.ToInt()] = to; // break the cycle.
forwarded = true;
} else {
TRACE(" fw %d -> %d (forward)\n", from.ToInt(), to.ToInt());
result[from.ToInt()] = to_to; // forward the block.
forwarded = true;
}
if (pop) stack.pop();
}
RpoNumber unvisited() { return RpoNumber::FromInt(-1); }
RpoNumber onstack() { return RpoNumber::FromInt(-2); }
};
struct GapJumpRecord {
GapJumpRecord(Zone* zone) : zone_(zone), gap_jump_records_(zone) {}
struct Record {
RpoNumber block;
Instruction* instr;
};
struct RpoNumberHash {
std::size_t operator()(const RpoNumber& key) const {
return std::hash<int>()(key.ToInt());
}
};
bool CanForwardGapJump(Instruction* instr, RpoNumber instr_block,
RpoNumber target_block, RpoNumber* forward_to) {
DCHECK_EQ(instr->arch_opcode(), kArchJmp);
bool can_forward = false;
auto search = gap_jump_records_.find(target_block);
if (search != gap_jump_records_.end()) {
for (Record& record : search->second) {
Instruction* record_instr = record.instr;
DCHECK_EQ(record_instr->arch_opcode(), kArchJmp);
bool is_same_instr = true;
for (int i = Instruction::FIRST_GAP_POSITION;
i <= Instruction::LAST_GAP_POSITION; i++) {
Instruction::GapPosition pos =
static_cast<Instruction::GapPosition>(i);
ParallelMove* record_move = record_instr->GetParallelMove(pos);
ParallelMove* instr_move = instr->GetParallelMove(pos);
if (record_move == nullptr && instr_move == nullptr) continue;
if (((record_move == nullptr) != (instr_move == nullptr)) ||
!record_move->Equals(*instr_move)) {
is_same_instr = false;
break;
}
}
if (is_same_instr) {
// Found an instruction same as the recorded one.
*forward_to = record.block;
can_forward = true;
break;
}
}
if (!can_forward) {
// No recorded instruction has been found for this target block,
// so create a new record with the given instruction.
search->second.push_back({instr_block, instr});
}
} else {
// This is the first explored gap jump to target block.
auto ins =
gap_jump_records_.insert({target_block, ZoneVector<Record>(zone_)});
if (ins.second) {
ins.first->second.reserve(4);
ins.first->second.push_back({instr_block, instr});
}
}
return can_forward;
}
Zone* zone_;
ZoneUnorderedMap<RpoNumber, ZoneVector<Record>, RpoNumberHash>
gap_jump_records_;
};
} // namespace
bool JumpThreading::ComputeForwarding(Zone* local_zone,
ZoneVector<RpoNumber>* result,
InstructionSequence* code,
bool frame_at_start) {
ZoneStack<RpoNumber> stack(local_zone);
JumpThreadingState state = {false, *result, stack};
state.Clear(code->InstructionBlockCount());
RpoNumber empty_deconstruct_frame_return_block = RpoNumber::Invalid();
int32_t empty_deconstruct_frame_return_size;
RpoNumber empty_no_deconstruct_frame_return_block = RpoNumber::Invalid();
int32_t empty_no_deconstruct_frame_return_size;
GapJumpRecord record(local_zone);
// Iterate over the blocks forward, pushing the blocks onto the stack.
for (auto const instruction_block : code->instruction_blocks()) {
RpoNumber current = instruction_block->rpo_number();
state.PushIfUnvisited(current);
// Process the stack, which implements DFS through empty blocks.
while (!state.stack.empty()) {
InstructionBlock* block = code->InstructionBlockAt(state.stack.top());
// Process the instructions in a block up to a non-empty instruction.
TRACE("jt [%d] B%d\n", static_cast<int>(stack.size()),
block->rpo_number().ToInt());
RpoNumber fw = block->rpo_number();
bool fallthru = true;
for (int i = block->code_start(); i < block->code_end(); ++i) {
Instruction* instr = code->InstructionAt(i);
if (!instr->AreMovesRedundant()) {
TRACE(" parallel move");
// can't skip instructions with non redundant moves, except when we
// can forward to a block with identical gap-moves.
if (instr->arch_opcode() == kArchJmp) {
TRACE(" jmp");
RpoNumber forward_to;
if ((frame_at_start || !(block->must_deconstruct_frame() ||
block->must_construct_frame())) &&
record.CanForwardGapJump(instr, block->rpo_number(),
code->InputRpo(instr, 0),
&forward_to)) {
DCHECK(forward_to.IsValid());
fw = forward_to;
TRACE("\n merge B%d into B%d", block->rpo_number().ToInt(),
forward_to.ToInt());
}
}
TRACE("\n");
fallthru = false;
} else if (FlagsModeField::decode(instr->opcode()) != kFlags_none) {
// can't skip instructions with flags continuations.
TRACE(" flags\n");
fallthru = false;
} else if (instr->IsNop()) {
// skip nops.
TRACE(" nop\n");
continue;
} else if (instr->arch_opcode() == kArchJmp) {
// try to forward the jump instruction.
TRACE(" jmp\n");
// if this block deconstructs the frame, we can't forward it.
// TODO(mtrofin): we can still forward if we end up building
// the frame at start. So we should move the decision of whether
// to build a frame or not in the register allocator, and trickle it
// here and to the code generator.
if (frame_at_start || !(block->must_deconstruct_frame() ||
block->must_construct_frame())) {
fw = code->InputRpo(instr, 0);
}
fallthru = false;
} else if (instr->IsRet()) {
TRACE(" ret\n");
if (fallthru) {
CHECK_IMPLIES(block->must_construct_frame(),
block->must_deconstruct_frame());
// Only handle returns with immediate/constant operands, since
// they must always be the same for all returns in a function.
// Dynamic return values might use different registers at
// different return sites and therefore cannot be shared.
if (instr->InputAt(0)->IsImmediate()) {
int32_t return_size = ImmediateOperand::cast(instr->InputAt(0))
->inline_int32_value();
// Instructions can be shared only for blocks that share
// the same |must_deconstruct_frame| attribute.
if (block->must_deconstruct_frame()) {
if (empty_deconstruct_frame_return_block ==
RpoNumber::Invalid()) {
empty_deconstruct_frame_return_block = block->rpo_number();
empty_deconstruct_frame_return_size = return_size;
} else if (empty_deconstruct_frame_return_size == return_size) {
fw = empty_deconstruct_frame_return_block;
block->clear_must_deconstruct_frame();
}
} else {
if (empty_no_deconstruct_frame_return_block ==
RpoNumber::Invalid()) {
empty_no_deconstruct_frame_return_block = block->rpo_number();
empty_no_deconstruct_frame_return_size = return_size;
} else if (empty_no_deconstruct_frame_return_size ==
return_size) {
fw = empty_no_deconstruct_frame_return_block;
}
}
}
}
fallthru = false;
} else {
// can't skip other instructions.
TRACE(" other\n");
fallthru = false;
}
break;
}
if (fallthru) {
int next = 1 + block->rpo_number().ToInt();
if (next < code->InstructionBlockCount()) fw = RpoNumber::FromInt(next);
}
state.Forward(fw);
}
}
#ifdef DEBUG
for (RpoNumber num : *result) {
DCHECK(num.IsValid());
}
#endif
if (v8_flags.trace_turbo_jt) {
for (int i = 0; i < static_cast<int>(result->size()); i++) {
TRACE("B%d ", i);
int to = (*result)[i].ToInt();
if (i != to) {
TRACE("-> B%d\n", to);
} else {
TRACE("\n");
}
}
}
return state.forwarded;
}
void JumpThreading::ApplyForwarding(Zone* local_zone,
ZoneVector<RpoNumber> const& result,
InstructionSequence* code) {
if (!v8_flags.turbo_jt) return;
ZoneVector<bool> skip(static_cast<int>(result.size()), false, local_zone);
// Skip empty blocks when the previous block doesn't fall through.
bool prev_fallthru = true;
for (auto const block : code->ao_blocks()) {
RpoNumber block_rpo = block->rpo_number();
int block_num = block_rpo.ToInt();
RpoNumber result_rpo = result[block_num];
skip[block_num] = !prev_fallthru && result_rpo != block_rpo;
if (result_rpo != block_rpo) {
// We need the handler information to be propagated, so that branch
// targets are annotated as necessary for control flow integrity
// checks (when enabled).
if (code->InstructionBlockAt(block_rpo)->IsHandler()) {
code->InstructionBlockAt(result_rpo)->MarkHandler();
}
}
bool fallthru = true;
for (int i = block->code_start(); i < block->code_end(); ++i) {
Instruction* instr = code->InstructionAt(i);
FlagsMode mode = FlagsModeField::decode(instr->opcode());
if (mode == kFlags_branch) {
fallthru = false; // branches don't fall through to the next block.
} else if (instr->arch_opcode() == kArchJmp ||
instr->arch_opcode() == kArchRet) {
if (skip[block_num]) {
// Overwrite a redundant jump with a nop.
TRACE("jt-fw nop @%d\n", i);
instr->OverwriteWithNop();
// Eliminate all the ParallelMoves.
for (int i = Instruction::FIRST_GAP_POSITION;
i <= Instruction::LAST_GAP_POSITION; i++) {
Instruction::GapPosition pos =
static_cast<Instruction::GapPosition>(i);
ParallelMove* instr_move = instr->GetParallelMove(pos);
if (instr_move != nullptr) {
instr_move->Eliminate();
}
}
// If this block was marked as a handler, it can be unmarked now.
code->InstructionBlockAt(block_rpo)->UnmarkHandler();
}
fallthru = false; // jumps don't fall through to the next block.
}
}
prev_fallthru = fallthru;
}
// Patch RPO immediates.
InstructionSequence::RpoImmediates& rpo_immediates = code->rpo_immediates();
for (size_t i = 0; i < rpo_immediates.size(); i++) {
RpoNumber rpo = rpo_immediates[i];
if (rpo.IsValid()) {
RpoNumber fw = result[rpo.ToInt()];
if (fw != rpo) rpo_immediates[i] = fw;
}
}
// Renumber the blocks so that IsNextInAssemblyOrder() will return true,
// even if there are skipped blocks in-between.
int ao = 0;
for (auto const block : code->ao_blocks()) {
block->set_ao_number(RpoNumber::FromInt(ao));
if (!skip[block->rpo_number().ToInt()]) ao++;
}
}
#undef TRACE
} // namespace compiler
} // namespace internal
} // namespace v8
|