Newer
Older
//===- X86InstrInfo.cpp - X86 Instruction Information -----------*- C++ -*-===//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//===----------------------------------------------------------------------===//
// This file contains the X86 implementation of the TargetInstrInfo class.
//
//===----------------------------------------------------------------------===//
#include "X86.h"
Chris Lattner
committed
#include "X86InstrBuilder.h"
#include "X86MachineFunctionInfo.h"
#include "X86Subtarget.h"
#include "X86TargetMachine.h"
#include "llvm/ADT/STLExtras.h"
#include "llvm/CodeGen/MachineFrameInfo.h"
Misha Brukman
committed
#include "llvm/CodeGen/MachineInstrBuilder.h"
#include "llvm/CodeGen/MachineRegisterInfo.h"
Evan Cheng
committed
#include "llvm/CodeGen/LiveVariables.h"
#include "llvm/Support/CommandLine.h"
Evan Cheng
committed
#include "llvm/Target/TargetOptions.h"
using namespace llvm;
namespace {
cl::opt<bool>
NoFusing("disable-spill-fusing",
cl::desc("Disable fusing of spill code into instructions"));
cl::opt<bool>
PrintFailedFusing("print-failed-fuse-candidates",
cl::desc("Print instructions that the allocator wants to"
" fuse, but the X86 backend currently can't"),
cl::Hidden);
}
: TargetInstrInfoImpl(X86Insts, array_lengthof(X86Insts)),
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
SmallVector<unsigned,16> AmbEntries;
static const unsigned OpTbl2Addr[][2] = {
{ X86::ADC32ri, X86::ADC32mi },
{ X86::ADC32ri8, X86::ADC32mi8 },
{ X86::ADC32rr, X86::ADC32mr },
{ X86::ADC64ri32, X86::ADC64mi32 },
{ X86::ADC64ri8, X86::ADC64mi8 },
{ X86::ADC64rr, X86::ADC64mr },
{ X86::ADD16ri, X86::ADD16mi },
{ X86::ADD16ri8, X86::ADD16mi8 },
{ X86::ADD16rr, X86::ADD16mr },
{ X86::ADD32ri, X86::ADD32mi },
{ X86::ADD32ri8, X86::ADD32mi8 },
{ X86::ADD32rr, X86::ADD32mr },
{ X86::ADD64ri32, X86::ADD64mi32 },
{ X86::ADD64ri8, X86::ADD64mi8 },
{ X86::ADD64rr, X86::ADD64mr },
{ X86::ADD8ri, X86::ADD8mi },
{ X86::ADD8rr, X86::ADD8mr },
{ X86::AND16ri, X86::AND16mi },
{ X86::AND16ri8, X86::AND16mi8 },
{ X86::AND16rr, X86::AND16mr },
{ X86::AND32ri, X86::AND32mi },
{ X86::AND32ri8, X86::AND32mi8 },
{ X86::AND32rr, X86::AND32mr },
{ X86::AND64ri32, X86::AND64mi32 },
{ X86::AND64ri8, X86::AND64mi8 },
{ X86::AND64rr, X86::AND64mr },
{ X86::AND8ri, X86::AND8mi },
{ X86::AND8rr, X86::AND8mr },
{ X86::DEC16r, X86::DEC16m },
{ X86::DEC32r, X86::DEC32m },
{ X86::DEC64_16r, X86::DEC64_16m },
{ X86::DEC64_32r, X86::DEC64_32m },
{ X86::DEC64r, X86::DEC64m },
{ X86::DEC8r, X86::DEC8m },
{ X86::INC16r, X86::INC16m },
{ X86::INC32r, X86::INC32m },
{ X86::INC64_16r, X86::INC64_16m },
{ X86::INC64_32r, X86::INC64_32m },
{ X86::INC64r, X86::INC64m },
{ X86::INC8r, X86::INC8m },
{ X86::NEG16r, X86::NEG16m },
{ X86::NEG32r, X86::NEG32m },
{ X86::NEG64r, X86::NEG64m },
{ X86::NEG8r, X86::NEG8m },
{ X86::NOT16r, X86::NOT16m },
{ X86::NOT32r, X86::NOT32m },
{ X86::NOT64r, X86::NOT64m },
{ X86::NOT8r, X86::NOT8m },
{ X86::OR16ri, X86::OR16mi },
{ X86::OR16ri8, X86::OR16mi8 },
{ X86::OR16rr, X86::OR16mr },
{ X86::OR32ri, X86::OR32mi },
{ X86::OR32ri8, X86::OR32mi8 },
{ X86::OR32rr, X86::OR32mr },
{ X86::OR64ri32, X86::OR64mi32 },
{ X86::OR64ri8, X86::OR64mi8 },
{ X86::OR64rr, X86::OR64mr },
{ X86::OR8ri, X86::OR8mi },
{ X86::OR8rr, X86::OR8mr },
{ X86::ROL16r1, X86::ROL16m1 },
{ X86::ROL16rCL, X86::ROL16mCL },
{ X86::ROL16ri, X86::ROL16mi },
{ X86::ROL32r1, X86::ROL32m1 },
{ X86::ROL32rCL, X86::ROL32mCL },
{ X86::ROL32ri, X86::ROL32mi },
{ X86::ROL64r1, X86::ROL64m1 },
{ X86::ROL64rCL, X86::ROL64mCL },
{ X86::ROL64ri, X86::ROL64mi },
{ X86::ROL8r1, X86::ROL8m1 },
{ X86::ROL8rCL, X86::ROL8mCL },
{ X86::ROL8ri, X86::ROL8mi },
{ X86::ROR16r1, X86::ROR16m1 },
{ X86::ROR16rCL, X86::ROR16mCL },
{ X86::ROR16ri, X86::ROR16mi },
{ X86::ROR32r1, X86::ROR32m1 },
{ X86::ROR32rCL, X86::ROR32mCL },
{ X86::ROR32ri, X86::ROR32mi },
{ X86::ROR64r1, X86::ROR64m1 },
{ X86::ROR64rCL, X86::ROR64mCL },
{ X86::ROR64ri, X86::ROR64mi },
{ X86::ROR8r1, X86::ROR8m1 },
{ X86::ROR8rCL, X86::ROR8mCL },
{ X86::ROR8ri, X86::ROR8mi },
{ X86::SAR16r1, X86::SAR16m1 },
{ X86::SAR16rCL, X86::SAR16mCL },
{ X86::SAR16ri, X86::SAR16mi },
{ X86::SAR32r1, X86::SAR32m1 },
{ X86::SAR32rCL, X86::SAR32mCL },
{ X86::SAR32ri, X86::SAR32mi },
{ X86::SAR64r1, X86::SAR64m1 },
{ X86::SAR64rCL, X86::SAR64mCL },
{ X86::SAR64ri, X86::SAR64mi },
{ X86::SAR8r1, X86::SAR8m1 },
{ X86::SAR8rCL, X86::SAR8mCL },
{ X86::SAR8ri, X86::SAR8mi },
{ X86::SBB32ri, X86::SBB32mi },
{ X86::SBB32ri8, X86::SBB32mi8 },
{ X86::SBB32rr, X86::SBB32mr },
{ X86::SBB64ri32, X86::SBB64mi32 },
{ X86::SBB64ri8, X86::SBB64mi8 },
{ X86::SBB64rr, X86::SBB64mr },
{ X86::SHL16rCL, X86::SHL16mCL },
{ X86::SHL16ri, X86::SHL16mi },
{ X86::SHL32rCL, X86::SHL32mCL },
{ X86::SHL32ri, X86::SHL32mi },
{ X86::SHL64rCL, X86::SHL64mCL },
{ X86::SHL64ri, X86::SHL64mi },
{ X86::SHL8rCL, X86::SHL8mCL },
{ X86::SHL8ri, X86::SHL8mi },
{ X86::SHLD16rrCL, X86::SHLD16mrCL },
{ X86::SHLD16rri8, X86::SHLD16mri8 },
{ X86::SHLD32rrCL, X86::SHLD32mrCL },
{ X86::SHLD32rri8, X86::SHLD32mri8 },
{ X86::SHLD64rrCL, X86::SHLD64mrCL },
{ X86::SHLD64rri8, X86::SHLD64mri8 },
{ X86::SHR16r1, X86::SHR16m1 },
{ X86::SHR16rCL, X86::SHR16mCL },
{ X86::SHR16ri, X86::SHR16mi },
{ X86::SHR32r1, X86::SHR32m1 },
{ X86::SHR32rCL, X86::SHR32mCL },
{ X86::SHR32ri, X86::SHR32mi },
{ X86::SHR64r1, X86::SHR64m1 },
{ X86::SHR64rCL, X86::SHR64mCL },
{ X86::SHR64ri, X86::SHR64mi },
{ X86::SHR8r1, X86::SHR8m1 },
{ X86::SHR8rCL, X86::SHR8mCL },
{ X86::SHR8ri, X86::SHR8mi },
{ X86::SHRD16rrCL, X86::SHRD16mrCL },
{ X86::SHRD16rri8, X86::SHRD16mri8 },
{ X86::SHRD32rrCL, X86::SHRD32mrCL },
{ X86::SHRD32rri8, X86::SHRD32mri8 },
{ X86::SHRD64rrCL, X86::SHRD64mrCL },
{ X86::SHRD64rri8, X86::SHRD64mri8 },
{ X86::SUB16ri, X86::SUB16mi },
{ X86::SUB16ri8, X86::SUB16mi8 },
{ X86::SUB16rr, X86::SUB16mr },
{ X86::SUB32ri, X86::SUB32mi },
{ X86::SUB32ri8, X86::SUB32mi8 },
{ X86::SUB32rr, X86::SUB32mr },
{ X86::SUB64ri32, X86::SUB64mi32 },
{ X86::SUB64ri8, X86::SUB64mi8 },
{ X86::SUB64rr, X86::SUB64mr },
{ X86::SUB8ri, X86::SUB8mi },
{ X86::SUB8rr, X86::SUB8mr },
{ X86::XOR16ri, X86::XOR16mi },
{ X86::XOR16ri8, X86::XOR16mi8 },
{ X86::XOR16rr, X86::XOR16mr },
{ X86::XOR32ri, X86::XOR32mi },
{ X86::XOR32ri8, X86::XOR32mi8 },
{ X86::XOR32rr, X86::XOR32mr },
{ X86::XOR64ri32, X86::XOR64mi32 },
{ X86::XOR64ri8, X86::XOR64mi8 },
{ X86::XOR64rr, X86::XOR64mr },
{ X86::XOR8ri, X86::XOR8mi },
{ X86::XOR8rr, X86::XOR8mr }
};
for (unsigned i = 0, e = array_lengthof(OpTbl2Addr); i != e; ++i) {
unsigned RegOp = OpTbl2Addr[i][0];
unsigned MemOp = OpTbl2Addr[i][1];
if (!RegOp2MemOpTable2Addr.insert(std::make_pair((unsigned*)RegOp, MemOp)))
assert(false && "Duplicated entries?");
unsigned AuxInfo = 0 | (1 << 4) | (1 << 5); // Index 0,folded load and store
if (!MemOp2RegOpTable.insert(std::make_pair((unsigned*)MemOp,
std::make_pair(RegOp, AuxInfo))))
AmbEntries.push_back(MemOp);
}
// If the third value is 1, then it's folding either a load or a store.
static const unsigned OpTbl0[][3] = {
{ X86::CALL32r, X86::CALL32m, 1 },
{ X86::CALL64r, X86::CALL64m, 1 },
{ X86::CMP16ri, X86::CMP16mi, 1 },
{ X86::CMP16ri8, X86::CMP16mi8, 1 },
{ X86::CMP16rr, X86::CMP16mr, 1 },
{ X86::CMP32ri, X86::CMP32mi, 1 },
{ X86::CMP32ri8, X86::CMP32mi8, 1 },
{ X86::CMP32rr, X86::CMP32mr, 1 },
{ X86::CMP64ri32, X86::CMP64mi32, 1 },
{ X86::CMP64ri8, X86::CMP64mi8, 1 },
{ X86::CMP64rr, X86::CMP64mr, 1 },
{ X86::CMP8ri, X86::CMP8mi, 1 },
{ X86::CMP8rr, X86::CMP8mr, 1 },
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
{ X86::DIV16r, X86::DIV16m, 1 },
{ X86::DIV32r, X86::DIV32m, 1 },
{ X86::DIV64r, X86::DIV64m, 1 },
{ X86::DIV8r, X86::DIV8m, 1 },
{ X86::FsMOVAPDrr, X86::MOVSDmr, 0 },
{ X86::FsMOVAPSrr, X86::MOVSSmr, 0 },
{ X86::IDIV16r, X86::IDIV16m, 1 },
{ X86::IDIV32r, X86::IDIV32m, 1 },
{ X86::IDIV64r, X86::IDIV64m, 1 },
{ X86::IDIV8r, X86::IDIV8m, 1 },
{ X86::IMUL16r, X86::IMUL16m, 1 },
{ X86::IMUL32r, X86::IMUL32m, 1 },
{ X86::IMUL64r, X86::IMUL64m, 1 },
{ X86::IMUL8r, X86::IMUL8m, 1 },
{ X86::JMP32r, X86::JMP32m, 1 },
{ X86::JMP64r, X86::JMP64m, 1 },
{ X86::MOV16ri, X86::MOV16mi, 0 },
{ X86::MOV16rr, X86::MOV16mr, 0 },
{ X86::MOV16to16_, X86::MOV16_mr, 0 },
{ X86::MOV32ri, X86::MOV32mi, 0 },
{ X86::MOV32rr, X86::MOV32mr, 0 },
{ X86::MOV32to32_, X86::MOV32_mr, 0 },
{ X86::MOV64ri32, X86::MOV64mi32, 0 },
{ X86::MOV64rr, X86::MOV64mr, 0 },
{ X86::MOV8ri, X86::MOV8mi, 0 },
{ X86::MOV8rr, X86::MOV8mr, 0 },
{ X86::MOVAPDrr, X86::MOVAPDmr, 0 },
{ X86::MOVAPSrr, X86::MOVAPSmr, 0 },
{ X86::MOVPDI2DIrr, X86::MOVPDI2DImr, 0 },
{ X86::MOVPQIto64rr,X86::MOVPQI2QImr, 0 },
{ X86::MOVPS2SSrr, X86::MOVPS2SSmr, 0 },
{ X86::MOVSDrr, X86::MOVSDmr, 0 },
{ X86::MOVSDto64rr, X86::MOVSDto64mr, 0 },
{ X86::MOVSS2DIrr, X86::MOVSS2DImr, 0 },
{ X86::MOVSSrr, X86::MOVSSmr, 0 },
{ X86::MOVUPDrr, X86::MOVUPDmr, 0 },
{ X86::MOVUPSrr, X86::MOVUPSmr, 0 },
{ X86::MUL16r, X86::MUL16m, 1 },
{ X86::MUL32r, X86::MUL32m, 1 },
{ X86::MUL64r, X86::MUL64m, 1 },
{ X86::MUL8r, X86::MUL8m, 1 },
{ X86::SETAEr, X86::SETAEm, 0 },
{ X86::SETAr, X86::SETAm, 0 },
{ X86::SETBEr, X86::SETBEm, 0 },
{ X86::SETBr, X86::SETBm, 0 },
{ X86::SETEr, X86::SETEm, 0 },
{ X86::SETGEr, X86::SETGEm, 0 },
{ X86::SETGr, X86::SETGm, 0 },
{ X86::SETLEr, X86::SETLEm, 0 },
{ X86::SETLr, X86::SETLm, 0 },
{ X86::SETNEr, X86::SETNEm, 0 },
{ X86::SETNPr, X86::SETNPm, 0 },
{ X86::SETNSr, X86::SETNSm, 0 },
{ X86::SETPr, X86::SETPm, 0 },
{ X86::SETSr, X86::SETSm, 0 },
{ X86::TAILJMPr, X86::TAILJMPm, 1 },
{ X86::TEST16ri, X86::TEST16mi, 1 },
{ X86::TEST32ri, X86::TEST32mi, 1 },
{ X86::TEST64ri32, X86::TEST64mi32, 1 },
{ X86::TEST8ri, X86::TEST8mi, 1 }
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
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
};
for (unsigned i = 0, e = array_lengthof(OpTbl0); i != e; ++i) {
unsigned RegOp = OpTbl0[i][0];
unsigned MemOp = OpTbl0[i][1];
if (!RegOp2MemOpTable0.insert(std::make_pair((unsigned*)RegOp, MemOp)))
assert(false && "Duplicated entries?");
unsigned FoldedLoad = OpTbl0[i][2];
// Index 0, folded load or store.
unsigned AuxInfo = 0 | (FoldedLoad << 4) | ((FoldedLoad^1) << 5);
if (RegOp != X86::FsMOVAPDrr && RegOp != X86::FsMOVAPSrr)
if (!MemOp2RegOpTable.insert(std::make_pair((unsigned*)MemOp,
std::make_pair(RegOp, AuxInfo))))
AmbEntries.push_back(MemOp);
}
static const unsigned OpTbl1[][2] = {
{ X86::CMP16rr, X86::CMP16rm },
{ X86::CMP32rr, X86::CMP32rm },
{ X86::CMP64rr, X86::CMP64rm },
{ X86::CMP8rr, X86::CMP8rm },
{ X86::CVTSD2SSrr, X86::CVTSD2SSrm },
{ X86::CVTSI2SD64rr, X86::CVTSI2SD64rm },
{ X86::CVTSI2SDrr, X86::CVTSI2SDrm },
{ X86::CVTSI2SS64rr, X86::CVTSI2SS64rm },
{ X86::CVTSI2SSrr, X86::CVTSI2SSrm },
{ X86::CVTSS2SDrr, X86::CVTSS2SDrm },
{ X86::CVTTSD2SI64rr, X86::CVTTSD2SI64rm },
{ X86::CVTTSD2SIrr, X86::CVTTSD2SIrm },
{ X86::CVTTSS2SI64rr, X86::CVTTSS2SI64rm },
{ X86::CVTTSS2SIrr, X86::CVTTSS2SIrm },
{ X86::FsMOVAPDrr, X86::MOVSDrm },
{ X86::FsMOVAPSrr, X86::MOVSSrm },
{ X86::IMUL16rri, X86::IMUL16rmi },
{ X86::IMUL16rri8, X86::IMUL16rmi8 },
{ X86::IMUL32rri, X86::IMUL32rmi },
{ X86::IMUL32rri8, X86::IMUL32rmi8 },
{ X86::IMUL64rri32, X86::IMUL64rmi32 },
{ X86::IMUL64rri8, X86::IMUL64rmi8 },
{ X86::Int_CMPSDrr, X86::Int_CMPSDrm },
{ X86::Int_CMPSSrr, X86::Int_CMPSSrm },
{ X86::Int_COMISDrr, X86::Int_COMISDrm },
{ X86::Int_COMISSrr, X86::Int_COMISSrm },
{ X86::Int_CVTDQ2PDrr, X86::Int_CVTDQ2PDrm },
{ X86::Int_CVTDQ2PSrr, X86::Int_CVTDQ2PSrm },
{ X86::Int_CVTPD2DQrr, X86::Int_CVTPD2DQrm },
{ X86::Int_CVTPD2PSrr, X86::Int_CVTPD2PSrm },
{ X86::Int_CVTPS2DQrr, X86::Int_CVTPS2DQrm },
{ X86::Int_CVTPS2PDrr, X86::Int_CVTPS2PDrm },
{ X86::Int_CVTSD2SI64rr,X86::Int_CVTSD2SI64rm },
{ X86::Int_CVTSD2SIrr, X86::Int_CVTSD2SIrm },
{ X86::Int_CVTSD2SSrr, X86::Int_CVTSD2SSrm },
{ X86::Int_CVTSI2SD64rr,X86::Int_CVTSI2SD64rm },
{ X86::Int_CVTSI2SDrr, X86::Int_CVTSI2SDrm },
{ X86::Int_CVTSI2SS64rr,X86::Int_CVTSI2SS64rm },
{ X86::Int_CVTSI2SSrr, X86::Int_CVTSI2SSrm },
{ X86::Int_CVTSS2SDrr, X86::Int_CVTSS2SDrm },
{ X86::Int_CVTSS2SI64rr,X86::Int_CVTSS2SI64rm },
{ X86::Int_CVTSS2SIrr, X86::Int_CVTSS2SIrm },
{ X86::Int_CVTTPD2DQrr, X86::Int_CVTTPD2DQrm },
{ X86::Int_CVTTPS2DQrr, X86::Int_CVTTPS2DQrm },
{ X86::Int_CVTTSD2SI64rr,X86::Int_CVTTSD2SI64rm },
{ X86::Int_CVTTSD2SIrr, X86::Int_CVTTSD2SIrm },
{ X86::Int_CVTTSS2SI64rr,X86::Int_CVTTSS2SI64rm },
{ X86::Int_CVTTSS2SIrr, X86::Int_CVTTSS2SIrm },
{ X86::Int_UCOMISDrr, X86::Int_UCOMISDrm },
{ X86::Int_UCOMISSrr, X86::Int_UCOMISSrm },
{ X86::MOV16rr, X86::MOV16rm },
{ X86::MOV16to16_, X86::MOV16_rm },
{ X86::MOV32rr, X86::MOV32rm },
{ X86::MOV32to32_, X86::MOV32_rm },
{ X86::MOV64rr, X86::MOV64rm },
{ X86::MOV64toPQIrr, X86::MOVQI2PQIrm },
{ X86::MOV64toSDrr, X86::MOV64toSDrm },
{ X86::MOV8rr, X86::MOV8rm },
{ X86::MOVAPDrr, X86::MOVAPDrm },
{ X86::MOVAPSrr, X86::MOVAPSrm },
{ X86::MOVDDUPrr, X86::MOVDDUPrm },
{ X86::MOVDI2PDIrr, X86::MOVDI2PDIrm },
{ X86::MOVDI2SSrr, X86::MOVDI2SSrm },
{ X86::MOVSD2PDrr, X86::MOVSD2PDrm },
{ X86::MOVSDrr, X86::MOVSDrm },
{ X86::MOVSHDUPrr, X86::MOVSHDUPrm },
{ X86::MOVSLDUPrr, X86::MOVSLDUPrm },
{ X86::MOVSS2PSrr, X86::MOVSS2PSrm },
{ X86::MOVSSrr, X86::MOVSSrm },
{ X86::MOVSX16rr8, X86::MOVSX16rm8 },
{ X86::MOVSX32rr16, X86::MOVSX32rm16 },
{ X86::MOVSX32rr8, X86::MOVSX32rm8 },
{ X86::MOVSX64rr16, X86::MOVSX64rm16 },
{ X86::MOVSX64rr32, X86::MOVSX64rm32 },
{ X86::MOVSX64rr8, X86::MOVSX64rm8 },
{ X86::MOVUPDrr, X86::MOVUPDrm },
{ X86::MOVUPSrr, X86::MOVUPSrm },
{ X86::MOVZDI2PDIrr, X86::MOVZDI2PDIrm },
{ X86::MOVZQI2PQIrr, X86::MOVZQI2PQIrm },
{ X86::MOVZPQILo2PQIrr, X86::MOVZPQILo2PQIrm },
{ X86::MOVZX16rr8, X86::MOVZX16rm8 },
{ X86::MOVZX32rr16, X86::MOVZX32rm16 },
{ X86::MOVZX32rr8, X86::MOVZX32rm8 },
{ X86::MOVZX64rr16, X86::MOVZX64rm16 },
{ X86::MOVZX64rr8, X86::MOVZX64rm8 },
{ X86::PSHUFDri, X86::PSHUFDmi },
{ X86::PSHUFHWri, X86::PSHUFHWmi },
{ X86::PSHUFLWri, X86::PSHUFLWmi },
{ X86::RCPPSr, X86::RCPPSm },
{ X86::RCPPSr_Int, X86::RCPPSm_Int },
{ X86::RSQRTPSr, X86::RSQRTPSm },
{ X86::RSQRTPSr_Int, X86::RSQRTPSm_Int },
{ X86::RSQRTSSr, X86::RSQRTSSm },
{ X86::RSQRTSSr_Int, X86::RSQRTSSm_Int },
{ X86::SQRTPDr, X86::SQRTPDm },
{ X86::SQRTPDr_Int, X86::SQRTPDm_Int },
{ X86::SQRTPSr, X86::SQRTPSm },
{ X86::SQRTPSr_Int, X86::SQRTPSm_Int },
{ X86::SQRTSDr, X86::SQRTSDm },
{ X86::SQRTSDr_Int, X86::SQRTSDm_Int },
{ X86::SQRTSSr, X86::SQRTSSm },
{ X86::SQRTSSr_Int, X86::SQRTSSm_Int },
{ X86::TEST16rr, X86::TEST16rm },
{ X86::TEST32rr, X86::TEST32rm },
{ X86::TEST64rr, X86::TEST64rm },
{ X86::TEST8rr, X86::TEST8rm },
// FIXME: TEST*rr EAX,EAX ---> CMP [mem], 0
{ X86::UCOMISDrr, X86::UCOMISDrm },
{ X86::UCOMISSrr, X86::UCOMISSrm }
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
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
};
for (unsigned i = 0, e = array_lengthof(OpTbl1); i != e; ++i) {
unsigned RegOp = OpTbl1[i][0];
unsigned MemOp = OpTbl1[i][1];
if (!RegOp2MemOpTable1.insert(std::make_pair((unsigned*)RegOp, MemOp)))
assert(false && "Duplicated entries?");
unsigned AuxInfo = 1 | (1 << 4); // Index 1, folded load
if (RegOp != X86::FsMOVAPDrr && RegOp != X86::FsMOVAPSrr)
if (!MemOp2RegOpTable.insert(std::make_pair((unsigned*)MemOp,
std::make_pair(RegOp, AuxInfo))))
AmbEntries.push_back(MemOp);
}
static const unsigned OpTbl2[][2] = {
{ X86::ADC32rr, X86::ADC32rm },
{ X86::ADC64rr, X86::ADC64rm },
{ X86::ADD16rr, X86::ADD16rm },
{ X86::ADD32rr, X86::ADD32rm },
{ X86::ADD64rr, X86::ADD64rm },
{ X86::ADD8rr, X86::ADD8rm },
{ X86::ADDPDrr, X86::ADDPDrm },
{ X86::ADDPSrr, X86::ADDPSrm },
{ X86::ADDSDrr, X86::ADDSDrm },
{ X86::ADDSSrr, X86::ADDSSrm },
{ X86::ADDSUBPDrr, X86::ADDSUBPDrm },
{ X86::ADDSUBPSrr, X86::ADDSUBPSrm },
{ X86::AND16rr, X86::AND16rm },
{ X86::AND32rr, X86::AND32rm },
{ X86::AND64rr, X86::AND64rm },
{ X86::AND8rr, X86::AND8rm },
{ X86::ANDNPDrr, X86::ANDNPDrm },
{ X86::ANDNPSrr, X86::ANDNPSrm },
{ X86::ANDPDrr, X86::ANDPDrm },
{ X86::ANDPSrr, X86::ANDPSrm },
{ X86::CMOVA16rr, X86::CMOVA16rm },
{ X86::CMOVA32rr, X86::CMOVA32rm },
{ X86::CMOVA64rr, X86::CMOVA64rm },
{ X86::CMOVAE16rr, X86::CMOVAE16rm },
{ X86::CMOVAE32rr, X86::CMOVAE32rm },
{ X86::CMOVAE64rr, X86::CMOVAE64rm },
{ X86::CMOVB16rr, X86::CMOVB16rm },
{ X86::CMOVB32rr, X86::CMOVB32rm },
{ X86::CMOVB64rr, X86::CMOVB64rm },
{ X86::CMOVBE16rr, X86::CMOVBE16rm },
{ X86::CMOVBE32rr, X86::CMOVBE32rm },
{ X86::CMOVBE64rr, X86::CMOVBE64rm },
{ X86::CMOVE16rr, X86::CMOVE16rm },
{ X86::CMOVE32rr, X86::CMOVE32rm },
{ X86::CMOVE64rr, X86::CMOVE64rm },
{ X86::CMOVG16rr, X86::CMOVG16rm },
{ X86::CMOVG32rr, X86::CMOVG32rm },
{ X86::CMOVG64rr, X86::CMOVG64rm },
{ X86::CMOVGE16rr, X86::CMOVGE16rm },
{ X86::CMOVGE32rr, X86::CMOVGE32rm },
{ X86::CMOVGE64rr, X86::CMOVGE64rm },
{ X86::CMOVL16rr, X86::CMOVL16rm },
{ X86::CMOVL32rr, X86::CMOVL32rm },
{ X86::CMOVL64rr, X86::CMOVL64rm },
{ X86::CMOVLE16rr, X86::CMOVLE16rm },
{ X86::CMOVLE32rr, X86::CMOVLE32rm },
{ X86::CMOVLE64rr, X86::CMOVLE64rm },
{ X86::CMOVNE16rr, X86::CMOVNE16rm },
{ X86::CMOVNE32rr, X86::CMOVNE32rm },
{ X86::CMOVNE64rr, X86::CMOVNE64rm },
{ X86::CMOVNP16rr, X86::CMOVNP16rm },
{ X86::CMOVNP32rr, X86::CMOVNP32rm },
{ X86::CMOVNP64rr, X86::CMOVNP64rm },
{ X86::CMOVNS16rr, X86::CMOVNS16rm },
{ X86::CMOVNS32rr, X86::CMOVNS32rm },
{ X86::CMOVNS64rr, X86::CMOVNS64rm },
{ X86::CMOVP16rr, X86::CMOVP16rm },
{ X86::CMOVP32rr, X86::CMOVP32rm },
{ X86::CMOVP64rr, X86::CMOVP64rm },
{ X86::CMOVS16rr, X86::CMOVS16rm },
{ X86::CMOVS32rr, X86::CMOVS32rm },
{ X86::CMOVS64rr, X86::CMOVS64rm },
{ X86::CMPPDrri, X86::CMPPDrmi },
{ X86::CMPPSrri, X86::CMPPSrmi },
{ X86::CMPSDrr, X86::CMPSDrm },
{ X86::CMPSSrr, X86::CMPSSrm },
{ X86::DIVPDrr, X86::DIVPDrm },
{ X86::DIVPSrr, X86::DIVPSrm },
{ X86::DIVSDrr, X86::DIVSDrm },
{ X86::DIVSSrr, X86::DIVSSrm },
{ X86::FsANDNPDrr, X86::FsANDNPDrm },
{ X86::FsANDNPSrr, X86::FsANDNPSrm },
{ X86::FsANDPDrr, X86::FsANDPDrm },
{ X86::FsANDPSrr, X86::FsANDPSrm },
{ X86::FsORPDrr, X86::FsORPDrm },
{ X86::FsORPSrr, X86::FsORPSrm },
{ X86::FsXORPDrr, X86::FsXORPDrm },
{ X86::FsXORPSrr, X86::FsXORPSrm },
509
510
511
512
513
514
515
516
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
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
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
{ X86::HADDPDrr, X86::HADDPDrm },
{ X86::HADDPSrr, X86::HADDPSrm },
{ X86::HSUBPDrr, X86::HSUBPDrm },
{ X86::HSUBPSrr, X86::HSUBPSrm },
{ X86::IMUL16rr, X86::IMUL16rm },
{ X86::IMUL32rr, X86::IMUL32rm },
{ X86::IMUL64rr, X86::IMUL64rm },
{ X86::MAXPDrr, X86::MAXPDrm },
{ X86::MAXPDrr_Int, X86::MAXPDrm_Int },
{ X86::MAXPSrr, X86::MAXPSrm },
{ X86::MAXPSrr_Int, X86::MAXPSrm_Int },
{ X86::MAXSDrr, X86::MAXSDrm },
{ X86::MAXSDrr_Int, X86::MAXSDrm_Int },
{ X86::MAXSSrr, X86::MAXSSrm },
{ X86::MAXSSrr_Int, X86::MAXSSrm_Int },
{ X86::MINPDrr, X86::MINPDrm },
{ X86::MINPDrr_Int, X86::MINPDrm_Int },
{ X86::MINPSrr, X86::MINPSrm },
{ X86::MINPSrr_Int, X86::MINPSrm_Int },
{ X86::MINSDrr, X86::MINSDrm },
{ X86::MINSDrr_Int, X86::MINSDrm_Int },
{ X86::MINSSrr, X86::MINSSrm },
{ X86::MINSSrr_Int, X86::MINSSrm_Int },
{ X86::MULPDrr, X86::MULPDrm },
{ X86::MULPSrr, X86::MULPSrm },
{ X86::MULSDrr, X86::MULSDrm },
{ X86::MULSSrr, X86::MULSSrm },
{ X86::OR16rr, X86::OR16rm },
{ X86::OR32rr, X86::OR32rm },
{ X86::OR64rr, X86::OR64rm },
{ X86::OR8rr, X86::OR8rm },
{ X86::ORPDrr, X86::ORPDrm },
{ X86::ORPSrr, X86::ORPSrm },
{ X86::PACKSSDWrr, X86::PACKSSDWrm },
{ X86::PACKSSWBrr, X86::PACKSSWBrm },
{ X86::PACKUSWBrr, X86::PACKUSWBrm },
{ X86::PADDBrr, X86::PADDBrm },
{ X86::PADDDrr, X86::PADDDrm },
{ X86::PADDQrr, X86::PADDQrm },
{ X86::PADDSBrr, X86::PADDSBrm },
{ X86::PADDSWrr, X86::PADDSWrm },
{ X86::PADDWrr, X86::PADDWrm },
{ X86::PANDNrr, X86::PANDNrm },
{ X86::PANDrr, X86::PANDrm },
{ X86::PAVGBrr, X86::PAVGBrm },
{ X86::PAVGWrr, X86::PAVGWrm },
{ X86::PCMPEQBrr, X86::PCMPEQBrm },
{ X86::PCMPEQDrr, X86::PCMPEQDrm },
{ X86::PCMPEQWrr, X86::PCMPEQWrm },
{ X86::PCMPGTBrr, X86::PCMPGTBrm },
{ X86::PCMPGTDrr, X86::PCMPGTDrm },
{ X86::PCMPGTWrr, X86::PCMPGTWrm },
{ X86::PINSRWrri, X86::PINSRWrmi },
{ X86::PMADDWDrr, X86::PMADDWDrm },
{ X86::PMAXSWrr, X86::PMAXSWrm },
{ X86::PMAXUBrr, X86::PMAXUBrm },
{ X86::PMINSWrr, X86::PMINSWrm },
{ X86::PMINUBrr, X86::PMINUBrm },
{ X86::PMULHUWrr, X86::PMULHUWrm },
{ X86::PMULHWrr, X86::PMULHWrm },
{ X86::PMULLWrr, X86::PMULLWrm },
{ X86::PMULUDQrr, X86::PMULUDQrm },
{ X86::PORrr, X86::PORrm },
{ X86::PSADBWrr, X86::PSADBWrm },
{ X86::PSLLDrr, X86::PSLLDrm },
{ X86::PSLLQrr, X86::PSLLQrm },
{ X86::PSLLWrr, X86::PSLLWrm },
{ X86::PSRADrr, X86::PSRADrm },
{ X86::PSRAWrr, X86::PSRAWrm },
{ X86::PSRLDrr, X86::PSRLDrm },
{ X86::PSRLQrr, X86::PSRLQrm },
{ X86::PSRLWrr, X86::PSRLWrm },
{ X86::PSUBBrr, X86::PSUBBrm },
{ X86::PSUBDrr, X86::PSUBDrm },
{ X86::PSUBSBrr, X86::PSUBSBrm },
{ X86::PSUBSWrr, X86::PSUBSWrm },
{ X86::PSUBWrr, X86::PSUBWrm },
{ X86::PUNPCKHBWrr, X86::PUNPCKHBWrm },
{ X86::PUNPCKHDQrr, X86::PUNPCKHDQrm },
{ X86::PUNPCKHQDQrr, X86::PUNPCKHQDQrm },
{ X86::PUNPCKHWDrr, X86::PUNPCKHWDrm },
{ X86::PUNPCKLBWrr, X86::PUNPCKLBWrm },
{ X86::PUNPCKLDQrr, X86::PUNPCKLDQrm },
{ X86::PUNPCKLQDQrr, X86::PUNPCKLQDQrm },
{ X86::PUNPCKLWDrr, X86::PUNPCKLWDrm },
{ X86::PXORrr, X86::PXORrm },
{ X86::SBB32rr, X86::SBB32rm },
{ X86::SBB64rr, X86::SBB64rm },
{ X86::SHUFPDrri, X86::SHUFPDrmi },
{ X86::SHUFPSrri, X86::SHUFPSrmi },
{ X86::SUB16rr, X86::SUB16rm },
{ X86::SUB32rr, X86::SUB32rm },
{ X86::SUB64rr, X86::SUB64rm },
{ X86::SUB8rr, X86::SUB8rm },
{ X86::SUBPDrr, X86::SUBPDrm },
{ X86::SUBPSrr, X86::SUBPSrm },
{ X86::SUBSDrr, X86::SUBSDrm },
{ X86::SUBSSrr, X86::SUBSSrm },
// FIXME: TEST*rr -> swapped operand of TEST*mr.
{ X86::UNPCKHPDrr, X86::UNPCKHPDrm },
{ X86::UNPCKHPSrr, X86::UNPCKHPSrm },
{ X86::UNPCKLPDrr, X86::UNPCKLPDrm },
{ X86::UNPCKLPSrr, X86::UNPCKLPSrm },
{ X86::XOR16rr, X86::XOR16rm },
{ X86::XOR32rr, X86::XOR32rm },
{ X86::XOR64rr, X86::XOR64rm },
{ X86::XOR8rr, X86::XOR8rm },
{ X86::XORPDrr, X86::XORPDrm },
{ X86::XORPSrr, X86::XORPSrm }
};
for (unsigned i = 0, e = array_lengthof(OpTbl2); i != e; ++i) {
unsigned RegOp = OpTbl2[i][0];
unsigned MemOp = OpTbl2[i][1];
if (!RegOp2MemOpTable2.insert(std::make_pair((unsigned*)RegOp, MemOp)))
assert(false && "Duplicated entries?");
unsigned AuxInfo = 2 | (1 << 4); // Index 1, folded load
if (!MemOp2RegOpTable.insert(std::make_pair((unsigned*)MemOp,
std::make_pair(RegOp, AuxInfo))))
AmbEntries.push_back(MemOp);
}
// Remove ambiguous entries.
assert(AmbEntries.empty() && "Duplicated entries in unfolding maps?");
bool X86InstrInfo::isMoveInstr(const MachineInstr& MI,
unsigned& sourceReg,
unsigned& destReg) const {
switch (MI.getOpcode()) {
default:
return false;
case X86::MOV8rr:
case X86::MOV16rr:
case X86::MOV32rr:
case X86::MOV64rr:
case X86::MOV16to16_:
case X86::MOV32to32_:
case X86::MOVSSrr:
case X86::MOVSDrr:
// FP Stack register class copies
case X86::MOV_Fp3232: case X86::MOV_Fp6464: case X86::MOV_Fp8080:
case X86::MOV_Fp3264: case X86::MOV_Fp3280:
case X86::MOV_Fp6432: case X86::MOV_Fp8032:
case X86::FsMOVAPSrr:
case X86::FsMOVAPDrr:
case X86::MOVAPSrr:
case X86::MOVAPDrr:
case X86::MOVSS2PSrr:
case X86::MOVSD2PDrr:
case X86::MOVPS2SSrr:
case X86::MOVPD2SDrr:
case X86::MMX_MOVD64rr:
case X86::MMX_MOVQ64rr:
assert(MI.getNumOperands() >= 2 &&
MI.getOperand(0).isRegister() &&
MI.getOperand(1).isRegister() &&
"invalid register-register move instruction");
sourceReg = MI.getOperand(1).getReg();
destReg = MI.getOperand(0).getReg();
return true;
}
}
Chris Lattner
committed
unsigned X86InstrInfo::isLoadFromStackSlot(MachineInstr *MI,
int &FrameIndex) const {
switch (MI->getOpcode()) {
default: break;
case X86::MOV8rm:
case X86::MOV16rm:
Evan Cheng
committed
case X86::MOV16_rm:
Chris Lattner
committed
case X86::MOV32rm:
Evan Cheng
committed
case X86::MOV32_rm:
Chris Lattner
committed
case X86::MOVSSrm:
case X86::MOVSDrm:
case X86::MOVAPSrm:
case X86::MOVAPDrm:
case X86::MMX_MOVD64rm:
case X86::MMX_MOVQ64rm:
if (MI->getOperand(1).isFI() && MI->getOperand(2).isImm() &&
MI->getOperand(3).isReg() && MI->getOperand(4).isImm() &&
Chris Lattner
committed
MI->getOperand(2).getImm() == 1 &&
Chris Lattner
committed
MI->getOperand(3).getReg() == 0 &&
Chris Lattner
committed
MI->getOperand(4).getImm() == 0) {
FrameIndex = MI->getOperand(1).getIndex();
Chris Lattner
committed
return MI->getOperand(0).getReg();
}
break;
}
return 0;
}
unsigned X86InstrInfo::isStoreToStackSlot(MachineInstr *MI,
int &FrameIndex) const {
switch (MI->getOpcode()) {
default: break;
case X86::MOV8mr:
case X86::MOV16mr:
Evan Cheng
committed
case X86::MOV16_mr:
Chris Lattner
committed
case X86::MOV32mr:
Evan Cheng
committed
case X86::MOV32_mr:
Chris Lattner
committed
case X86::MOVSSmr:
case X86::MOVSDmr:
case X86::MOVAPSmr:
case X86::MOVAPDmr:
case X86::MMX_MOVD64mr:
case X86::MMX_MOVQ64mr:
if (MI->getOperand(0).isFI() && MI->getOperand(1).isImm() &&
MI->getOperand(2).isReg() && MI->getOperand(3).isImm() &&
Chris Lattner
committed
MI->getOperand(1).getImm() == 1 &&
Chris Lattner
committed
MI->getOperand(3).getImm() == 0) {
FrameIndex = MI->getOperand(0).getIndex();
Chris Lattner
committed
return MI->getOperand(4).getReg();
}
break;
}
return 0;
}
static bool regIsPICBase(MachineInstr *MI, unsigned BaseReg) {
MachineRegisterInfo &MRI = MI->getParent()->getParent()->getRegInfo();
bool isPICBase = false;
for (MachineRegisterInfo::def_iterator I = MRI.def_begin(BaseReg),
E = MRI.def_end(); I != E; ++I) {
MachineInstr *DefMI = I.getOperand().getParent();
if (DefMI->getOpcode() != X86::MOVPC32r)
return false;
assert(!isPICBase && "More than one PIC base?");
isPICBase = true;
}
return isPICBase;
}
bool X86InstrInfo::isReallyTriviallyReMaterializable(MachineInstr *MI) const {
Dan Gohman
committed
switch (MI->getOpcode()) {
default: break;
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
case X86::MOV8rm:
case X86::MOV16rm:
case X86::MOV16_rm:
case X86::MOV32rm:
case X86::MOV32_rm:
case X86::MOV64rm:
case X86::LD_Fp64m:
case X86::MOVSSrm:
case X86::MOVSDrm:
case X86::MOVAPSrm:
case X86::MOVAPDrm:
case X86::MMX_MOVD64rm:
case X86::MMX_MOVQ64rm: {
// Loads from constant pools are trivially rematerializable.
if (MI->getOperand(1).isReg() &&
MI->getOperand(2).isImm() &&
MI->getOperand(3).isReg() && MI->getOperand(3).getReg() == 0 &&
MI->getOperand(4).isCPI()) {
unsigned BaseReg = MI->getOperand(1).getReg();
if (BaseReg == 0)
return true;
// Allow re-materialization of PIC load.
MachineRegisterInfo &MRI = MI->getParent()->getParent()->getRegInfo();
bool isPICBase = false;
for (MachineRegisterInfo::def_iterator I = MRI.def_begin(BaseReg),
E = MRI.def_end(); I != E; ++I) {
MachineInstr *DefMI = I.getOperand().getParent();
if (DefMI->getOpcode() != X86::MOVPC32r)
return false;
assert(!isPICBase && "More than one PIC base?");
isPICBase = true;
}
return isPICBase;
}
return false;
}
case X86::LEA32r:
case X86::LEA64r: {
if (MI->getOperand(1).isReg() &&
MI->getOperand(2).isImm() &&
MI->getOperand(3).isReg() && MI->getOperand(3).getReg() == 0 &&
!MI->getOperand(4).isReg()) {
// lea fi#, lea GV, etc. are all rematerializable.
unsigned BaseReg = MI->getOperand(1).getReg();
if (BaseReg == 0)
return true;
// Allow re-materialization of lea PICBase + x.
return regIsPICBase(MI, BaseReg);
}
return false;
}
Dan Gohman
committed
}
// All other instructions marked M_REMATERIALIZABLE are always trivially
// rematerializable.
return true;
Dan Gohman
committed
}
/// isInvariantLoad - Return true if the specified instruction (which is marked
/// mayLoad) is loading from a location whose value is invariant across the
/// function. For example, loading a value from the constant pool or from
/// from the argument area of a function if it does not change. This should
/// only return true of *all* loads the instruction does are invariant (if it
/// does multiple loads).
bool X86InstrInfo::isInvariantLoad(MachineInstr *MI) const {
// This code cares about loads from three cases: constant pool entries,
// invariant argument slots, and global stubs. In order to handle these cases
// for all of the myriad of X86 instructions, we just scan for a CP/FI/GV
// operand and base our analysis on it. This is safe because the address of
// none of these three cases is ever used as anything other than a load base
// and X86 doesn't have any instructions that load from multiple places.
for (unsigned i = 0, e = MI->getNumOperands(); i != e; ++i) {
const MachineOperand &MO = MI->getOperand(i);
// Loads from constant pools are trivially invariant.
if (MO.isCPI())
if (MO.isGlobal()) {
if (TM.getSubtarget<X86Subtarget>().GVRequiresExtraLoad(MO.getGlobal(),
TM, false))
return true;
return false;
}
// If this is a load from an invariant stack slot, the load is a constant.
if (MO.isFI()) {
const MachineFrameInfo &MFI =
*MI->getParent()->getParent()->getFrameInfo();
int Idx = MO.getIndex();
return MFI.isFixedObjectIndex(Idx) && MFI.isImmutableObjectIndex(Idx);
}
Bill Wendling
committed
}
// All other instances of these instructions are presumed to have other
// issues.
Bill Wendling
committed
}
/// hasLiveCondCodeDef - True if MI has a condition code def, e.g. EFLAGS, that
/// is not marked dead.
static bool hasLiveCondCodeDef(MachineInstr *MI) {
for (unsigned i = 0, e = MI->getNumOperands(); i != e; ++i) {
MachineOperand &MO = MI->getOperand(i);
if (MO.isRegister() && MO.isDef() &&
MO.getReg() == X86::EFLAGS && !MO.isDead()) {
return true;
}
}
return false;
}
Chris Lattner
committed
/// convertToThreeAddress - This method must be implemented by targets that
/// set the M_CONVERTIBLE_TO_3_ADDR flag. When this flag is set, the target
/// may be able to convert a two-address instruction into a true
/// three-address instruction on demand. This allows the X86 target (for
/// example) to convert ADD and SHL instructions into LEA instructions if they
/// would require register copies due to two-addressness.
///
/// This method returns a null pointer if the transformation cannot be
/// performed, otherwise it returns the new instruction.
///
Evan Cheng
committed
MachineInstr *
X86InstrInfo::convertToThreeAddress(MachineFunction::iterator &MFI,
MachineBasicBlock::iterator &MBBI,
LiveVariables &LV) const {
MachineInstr *MI = MBBI;
Chris Lattner
committed
// All instructions input are two-addr instructions. Get the known operands.
unsigned Dest = MI->getOperand(0).getReg();
unsigned Src = MI->getOperand(1).getReg();
Evan Cheng
committed
// FIXME: 16-bit LEA's are really slow on Athlons, but not bad on P4's. When
// we have better subtarget support, enable the 16-bit LEA generation here.
Evan Cheng
committed
bool DisableLEA16 = true;
unsigned MIOpc = MI->getOpcode();
switch (MIOpc) {
case X86::SHUFPSrri: {
assert(MI->getNumOperands() == 4 && "Unknown shufps instruction!");
if (!TM.getSubtarget<X86Subtarget>().hasSSE2()) return 0;
unsigned A = MI->getOperand(0).getReg();
unsigned B = MI->getOperand(1).getReg();
unsigned C = MI->getOperand(2).getReg();
unsigned M = MI->getOperand(3).getImm();
if (B != C) return 0;
NewMI = BuildMI(get(X86::PSHUFDri), A).addReg(B).addImm(M);
assert(MI->getNumOperands() >= 3 && "Unknown shift instruction!");
// NOTE: LEA doesn't produce flags like shift does, but LLVM never uses
// the flags produced by a shift yet, so this is safe.
unsigned Dest = MI->getOperand(0).getReg();
unsigned Src = MI->getOperand(1).getReg();
unsigned ShAmt = MI->getOperand(2).getImm();
if (ShAmt == 0 || ShAmt >= 4) return 0;
NewMI = BuildMI(get(X86::LEA64r), Dest)
.addReg(0).addImm(1 << ShAmt).addReg(Src).addImm(0);
break;
}
assert(MI->getNumOperands() >= 3 && "Unknown shift instruction!");
// NOTE: LEA doesn't produce flags like shift does, but LLVM never uses
// the flags produced by a shift yet, so this is safe.
unsigned Dest = MI->getOperand(0).getReg();
unsigned Src = MI->getOperand(1).getReg();
unsigned ShAmt = MI->getOperand(2).getImm();
if (ShAmt == 0 || ShAmt >= 4) return 0;
unsigned Opc = TM.getSubtarget<X86Subtarget>().is64Bit() ?
X86::LEA64_32r : X86::LEA32r;
NewMI = BuildMI(get(Opc), Dest)
.addReg(0).addImm(1 << ShAmt).addReg(Src).addImm(0);
break;
}
case X86::SHL16ri: {
assert(MI->getNumOperands() >= 3 && "Unknown shift instruction!");
// NOTE: LEA doesn't produce flags like shift does, but LLVM never uses
// the flags produced by a shift yet, so this is safe.
unsigned Dest = MI->getOperand(0).getReg();
unsigned Src = MI->getOperand(1).getReg();
unsigned ShAmt = MI->getOperand(2).getImm();
if (ShAmt == 0 || ShAmt >= 4) return 0;
Christopher Lamb
committed
if (DisableLEA16) {
// If 16-bit LEA is disabled, use 32-bit LEA via subregisters.
MachineRegisterInfo &RegInfo = MFI->getParent()->getRegInfo();
unsigned Opc = TM.getSubtarget<X86Subtarget>().is64Bit()
? X86::LEA64_32r : X86::LEA32r;
unsigned leaInReg = RegInfo.createVirtualRegister(&X86::GR32RegClass);
unsigned leaOutReg = RegInfo.createVirtualRegister(&X86::GR32RegClass);
// Build and insert into an implicit UNDEF value. This is OK because
// well be shifting and then extracting the lower 16-bits.
Christopher Lamb
committed
MachineInstr *Undef = BuildMI(get(X86::IMPLICIT_DEF), leaInReg);
Christopher Lamb
committed
BuildMI(get(X86::INSERT_SUBREG),leaInReg)
Christopher Lamb
committed
.addReg(leaInReg).addReg(Src).addImm(X86::SUBREG_16BIT);
Christopher Lamb
committed
NewMI = BuildMI(get(Opc), leaOutReg)
.addReg(0).addImm(1 << ShAmt).addReg(leaInReg).addImm(0);
MachineInstr *Ext =
BuildMI(get(X86::EXTRACT_SUBREG), Dest)
.addReg(leaOutReg).addImm(X86::SUBREG_16BIT);
Christopher Lamb
committed
Ext->copyKillDeadInfo(MI);
Christopher Lamb
committed
MFI->insert(MBBI, Undef);
Christopher Lamb
committed
MFI->insert(MBBI, Ins); // Insert the insert_subreg
LV.instructionChanged(MI, NewMI); // Update live variables
LV.addVirtualRegisterKilled(leaInReg, NewMI);
MFI->insert(MBBI, NewMI); // Insert the new inst
LV.addVirtualRegisterKilled(leaOutReg, Ext);
MFI->insert(MBBI, Ext); // Insert the extract_subreg
Christopher Lamb
committed
return Ext;
} else {
NewMI = BuildMI(get(X86::LEA16r), Dest)
.addReg(0).addImm(1 << ShAmt).addReg(Src).addImm(0);
}
}
default: {
// The following opcodes also sets the condition code register(s). Only
// convert them to equivalent lea if the condition code register def's
// are dead!
if (hasLiveCondCodeDef(MI))
return 0;
bool is64Bit = TM.getSubtarget<X86Subtarget>().is64Bit();
switch (MIOpc) {
default: return 0;
case X86::INC64r:
case X86::INC32r: {
assert(MI->getNumOperands() >= 2 && "Unknown inc instruction!");
unsigned Opc = MIOpc == X86::INC64r ? X86::LEA64r
: (is64Bit ? X86::LEA64_32r : X86::LEA32r);
NewMI = addRegOffset(BuildMI(get(Opc), Dest), Src, 1);
break;
}
case X86::INC16r:
case X86::INC64_16r:
if (DisableLEA16) return 0;