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
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
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
501
502
503
504
505
506
507
508
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
633
634
635
636
637
638
639
640
641
642
643
644
645
646
|
//===- X86RegisterInfo.td - Describe the X86 Register File --*- tablegen -*-==//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
// This file describes the X86 Register file, defining the registers themselves,
// aliases between the registers, and the register classes built out of the
// registers.
//
//===----------------------------------------------------------------------===//
class X86Reg<string n, bits<16> Enc, list<Register> subregs = []> : Register<n> {
let Namespace = "X86";
let HWEncoding = Enc;
let SubRegs = subregs;
}
// Subregister indices.
let Namespace = "X86" in {
def sub_8bit : SubRegIndex<8>;
def sub_8bit_hi : SubRegIndex<8, 8>;
def sub_8bit_hi_phony : SubRegIndex<8, 8>;
def sub_16bit : SubRegIndex<16>;
def sub_16bit_hi : SubRegIndex<16, 16>;
def sub_32bit : SubRegIndex<32>;
def sub_xmm : SubRegIndex<128>;
def sub_ymm : SubRegIndex<256>;
def sub_mask_0 : SubRegIndex<-1>;
def sub_mask_1 : SubRegIndex<-1, -1>;
}
//===----------------------------------------------------------------------===//
// Register definitions...
//
// In the register alias definitions below, we define which registers alias
// which others. We only specify which registers the small registers alias,
// because the register file generator is smart enough to figure out that
// AL aliases AX if we tell it that AX aliased AL (for example).
// Dwarf numbering is different for 32-bit and 64-bit, and there are
// variations by target as well. Currently the first entry is for X86-64,
// second - for EH on X86-32/Darwin and third is 'generic' one (X86-32/Linux
// and debug information on X86-32/Darwin)
// 8-bit registers
// Low registers
def AL : X86Reg<"al", 0>;
def DL : X86Reg<"dl", 2>;
def CL : X86Reg<"cl", 1>;
def BL : X86Reg<"bl", 3>;
// High registers. On x86-64, these cannot be used in any instruction
// with a REX prefix.
def AH : X86Reg<"ah", 4>;
def DH : X86Reg<"dh", 6>;
def CH : X86Reg<"ch", 5>;
def BH : X86Reg<"bh", 7>;
// X86-64 only, requires REX.
let CostPerUse = 1 in {
def SIL : X86Reg<"sil", 6>;
def DIL : X86Reg<"dil", 7>;
def BPL : X86Reg<"bpl", 5>;
def SPL : X86Reg<"spl", 4>;
def R8B : X86Reg<"r8b", 8>;
def R9B : X86Reg<"r9b", 9>;
def R10B : X86Reg<"r10b", 10>;
def R11B : X86Reg<"r11b", 11>;
def R12B : X86Reg<"r12b", 12>;
def R13B : X86Reg<"r13b", 13>;
def R14B : X86Reg<"r14b", 14>;
def R15B : X86Reg<"r15b", 15>;
}
let isArtificial = 1 in {
// High byte of the low 16 bits of the super-register:
def SIH : X86Reg<"", -1>;
def DIH : X86Reg<"", -1>;
def BPH : X86Reg<"", -1>;
def SPH : X86Reg<"", -1>;
def R8BH : X86Reg<"", -1>;
def R9BH : X86Reg<"", -1>;
def R10BH : X86Reg<"", -1>;
def R11BH : X86Reg<"", -1>;
def R12BH : X86Reg<"", -1>;
def R13BH : X86Reg<"", -1>;
def R14BH : X86Reg<"", -1>;
def R15BH : X86Reg<"", -1>;
// High word of the low 32 bits of the super-register:
def HAX : X86Reg<"", -1>;
def HDX : X86Reg<"", -1>;
def HCX : X86Reg<"", -1>;
def HBX : X86Reg<"", -1>;
def HSI : X86Reg<"", -1>;
def HDI : X86Reg<"", -1>;
def HBP : X86Reg<"", -1>;
def HSP : X86Reg<"", -1>;
def HIP : X86Reg<"", -1>;
def R8WH : X86Reg<"", -1>;
def R9WH : X86Reg<"", -1>;
def R10WH : X86Reg<"", -1>;
def R11WH : X86Reg<"", -1>;
def R12WH : X86Reg<"", -1>;
def R13WH : X86Reg<"", -1>;
def R14WH : X86Reg<"", -1>;
def R15WH : X86Reg<"", -1>;
}
// 16-bit registers
let SubRegIndices = [sub_8bit, sub_8bit_hi], CoveredBySubRegs = 1 in {
def AX : X86Reg<"ax", 0, [AL,AH]>;
def DX : X86Reg<"dx", 2, [DL,DH]>;
def CX : X86Reg<"cx", 1, [CL,CH]>;
def BX : X86Reg<"bx", 3, [BL,BH]>;
}
let SubRegIndices = [sub_8bit, sub_8bit_hi_phony], CoveredBySubRegs = 1 in {
def SI : X86Reg<"si", 6, [SIL,SIH]>;
def DI : X86Reg<"di", 7, [DIL,DIH]>;
def BP : X86Reg<"bp", 5, [BPL,BPH]>;
def SP : X86Reg<"sp", 4, [SPL,SPH]>;
}
def IP : X86Reg<"ip", 0>;
// X86-64 only, requires REX.
let SubRegIndices = [sub_8bit, sub_8bit_hi_phony], CostPerUse = 1,
CoveredBySubRegs = 1 in {
def R8W : X86Reg<"r8w", 8, [R8B,R8BH]>;
def R9W : X86Reg<"r9w", 9, [R9B,R9BH]>;
def R10W : X86Reg<"r10w", 10, [R10B,R10BH]>;
def R11W : X86Reg<"r11w", 11, [R11B,R11BH]>;
def R12W : X86Reg<"r12w", 12, [R12B,R12BH]>;
def R13W : X86Reg<"r13w", 13, [R13B,R13BH]>;
def R14W : X86Reg<"r14w", 14, [R14B,R14BH]>;
def R15W : X86Reg<"r15w", 15, [R15B,R15BH]>;
}
// 32-bit registers
let SubRegIndices = [sub_16bit, sub_16bit_hi], CoveredBySubRegs = 1 in {
def EAX : X86Reg<"eax", 0, [AX, HAX]>, DwarfRegNum<[-2, 0, 0]>;
def EDX : X86Reg<"edx", 2, [DX, HDX]>, DwarfRegNum<[-2, 2, 2]>;
def ECX : X86Reg<"ecx", 1, [CX, HCX]>, DwarfRegNum<[-2, 1, 1]>;
def EBX : X86Reg<"ebx", 3, [BX, HBX]>, DwarfRegNum<[-2, 3, 3]>;
def ESI : X86Reg<"esi", 6, [SI, HSI]>, DwarfRegNum<[-2, 6, 6]>;
def EDI : X86Reg<"edi", 7, [DI, HDI]>, DwarfRegNum<[-2, 7, 7]>;
def EBP : X86Reg<"ebp", 5, [BP, HBP]>, DwarfRegNum<[-2, 4, 5]>;
def ESP : X86Reg<"esp", 4, [SP, HSP]>, DwarfRegNum<[-2, 5, 4]>;
def EIP : X86Reg<"eip", 0, [IP, HIP]>, DwarfRegNum<[-2, 8, 8]>;
}
// X86-64 only, requires REX
let SubRegIndices = [sub_16bit, sub_16bit_hi], CostPerUse = 1,
CoveredBySubRegs = 1 in {
def R8D : X86Reg<"r8d", 8, [R8W,R8WH]>;
def R9D : X86Reg<"r9d", 9, [R9W,R9WH]>;
def R10D : X86Reg<"r10d", 10, [R10W,R10WH]>;
def R11D : X86Reg<"r11d", 11, [R11W,R11WH]>;
def R12D : X86Reg<"r12d", 12, [R12W,R12WH]>;
def R13D : X86Reg<"r13d", 13, [R13W,R13WH]>;
def R14D : X86Reg<"r14d", 14, [R14W,R14WH]>;
def R15D : X86Reg<"r15d", 15, [R15W,R15WH]>;
}
// 64-bit registers, X86-64 only
let SubRegIndices = [sub_32bit] in {
def RAX : X86Reg<"rax", 0, [EAX]>, DwarfRegNum<[0, -2, -2]>;
def RDX : X86Reg<"rdx", 2, [EDX]>, DwarfRegNum<[1, -2, -2]>;
def RCX : X86Reg<"rcx", 1, [ECX]>, DwarfRegNum<[2, -2, -2]>;
def RBX : X86Reg<"rbx", 3, [EBX]>, DwarfRegNum<[3, -2, -2]>;
def RSI : X86Reg<"rsi", 6, [ESI]>, DwarfRegNum<[4, -2, -2]>;
def RDI : X86Reg<"rdi", 7, [EDI]>, DwarfRegNum<[5, -2, -2]>;
def RBP : X86Reg<"rbp", 5, [EBP]>, DwarfRegNum<[6, -2, -2]>;
def RSP : X86Reg<"rsp", 4, [ESP]>, DwarfRegNum<[7, -2, -2]>;
// These also require REX.
let CostPerUse = 1 in {
def R8 : X86Reg<"r8", 8, [R8D]>, DwarfRegNum<[ 8, -2, -2]>;
def R9 : X86Reg<"r9", 9, [R9D]>, DwarfRegNum<[ 9, -2, -2]>;
def R10 : X86Reg<"r10", 10, [R10D]>, DwarfRegNum<[10, -2, -2]>;
def R11 : X86Reg<"r11", 11, [R11D]>, DwarfRegNum<[11, -2, -2]>;
def R12 : X86Reg<"r12", 12, [R12D]>, DwarfRegNum<[12, -2, -2]>;
def R13 : X86Reg<"r13", 13, [R13D]>, DwarfRegNum<[13, -2, -2]>;
def R14 : X86Reg<"r14", 14, [R14D]>, DwarfRegNum<[14, -2, -2]>;
def R15 : X86Reg<"r15", 15, [R15D]>, DwarfRegNum<[15, -2, -2]>;
def RIP : X86Reg<"rip", 0, [EIP]>, DwarfRegNum<[16, -2, -2]>;
}}
// MMX Registers. These are actually aliased to ST0 .. ST7
def MM0 : X86Reg<"mm0", 0>, DwarfRegNum<[41, 29, 29]>;
def MM1 : X86Reg<"mm1", 1>, DwarfRegNum<[42, 30, 30]>;
def MM2 : X86Reg<"mm2", 2>, DwarfRegNum<[43, 31, 31]>;
def MM3 : X86Reg<"mm3", 3>, DwarfRegNum<[44, 32, 32]>;
def MM4 : X86Reg<"mm4", 4>, DwarfRegNum<[45, 33, 33]>;
def MM5 : X86Reg<"mm5", 5>, DwarfRegNum<[46, 34, 34]>;
def MM6 : X86Reg<"mm6", 6>, DwarfRegNum<[47, 35, 35]>;
def MM7 : X86Reg<"mm7", 7>, DwarfRegNum<[48, 36, 36]>;
// Pseudo Floating Point registers
def FP0 : X86Reg<"fp0", 0>;
def FP1 : X86Reg<"fp1", 0>;
def FP2 : X86Reg<"fp2", 0>;
def FP3 : X86Reg<"fp3", 0>;
def FP4 : X86Reg<"fp4", 0>;
def FP5 : X86Reg<"fp5", 0>;
def FP6 : X86Reg<"fp6", 0>;
def FP7 : X86Reg<"fp7", 0>;
// XMM Registers, used by the various SSE instruction set extensions.
def XMM0: X86Reg<"xmm0", 0>, DwarfRegNum<[17, 21, 21]>;
def XMM1: X86Reg<"xmm1", 1>, DwarfRegNum<[18, 22, 22]>;
def XMM2: X86Reg<"xmm2", 2>, DwarfRegNum<[19, 23, 23]>;
def XMM3: X86Reg<"xmm3", 3>, DwarfRegNum<[20, 24, 24]>;
def XMM4: X86Reg<"xmm4", 4>, DwarfRegNum<[21, 25, 25]>;
def XMM5: X86Reg<"xmm5", 5>, DwarfRegNum<[22, 26, 26]>;
def XMM6: X86Reg<"xmm6", 6>, DwarfRegNum<[23, 27, 27]>;
def XMM7: X86Reg<"xmm7", 7>, DwarfRegNum<[24, 28, 28]>;
// X86-64 only
let CostPerUse = 1 in {
def XMM8: X86Reg<"xmm8", 8>, DwarfRegNum<[25, -2, -2]>;
def XMM9: X86Reg<"xmm9", 9>, DwarfRegNum<[26, -2, -2]>;
def XMM10: X86Reg<"xmm10", 10>, DwarfRegNum<[27, -2, -2]>;
def XMM11: X86Reg<"xmm11", 11>, DwarfRegNum<[28, -2, -2]>;
def XMM12: X86Reg<"xmm12", 12>, DwarfRegNum<[29, -2, -2]>;
def XMM13: X86Reg<"xmm13", 13>, DwarfRegNum<[30, -2, -2]>;
def XMM14: X86Reg<"xmm14", 14>, DwarfRegNum<[31, -2, -2]>;
def XMM15: X86Reg<"xmm15", 15>, DwarfRegNum<[32, -2, -2]>;
def XMM16: X86Reg<"xmm16", 16>, DwarfRegNum<[67, -2, -2]>;
def XMM17: X86Reg<"xmm17", 17>, DwarfRegNum<[68, -2, -2]>;
def XMM18: X86Reg<"xmm18", 18>, DwarfRegNum<[69, -2, -2]>;
def XMM19: X86Reg<"xmm19", 19>, DwarfRegNum<[70, -2, -2]>;
def XMM20: X86Reg<"xmm20", 20>, DwarfRegNum<[71, -2, -2]>;
def XMM21: X86Reg<"xmm21", 21>, DwarfRegNum<[72, -2, -2]>;
def XMM22: X86Reg<"xmm22", 22>, DwarfRegNum<[73, -2, -2]>;
def XMM23: X86Reg<"xmm23", 23>, DwarfRegNum<[74, -2, -2]>;
def XMM24: X86Reg<"xmm24", 24>, DwarfRegNum<[75, -2, -2]>;
def XMM25: X86Reg<"xmm25", 25>, DwarfRegNum<[76, -2, -2]>;
def XMM26: X86Reg<"xmm26", 26>, DwarfRegNum<[77, -2, -2]>;
def XMM27: X86Reg<"xmm27", 27>, DwarfRegNum<[78, -2, -2]>;
def XMM28: X86Reg<"xmm28", 28>, DwarfRegNum<[79, -2, -2]>;
def XMM29: X86Reg<"xmm29", 29>, DwarfRegNum<[80, -2, -2]>;
def XMM30: X86Reg<"xmm30", 30>, DwarfRegNum<[81, -2, -2]>;
def XMM31: X86Reg<"xmm31", 31>, DwarfRegNum<[82, -2, -2]>;
} // CostPerUse
// YMM0-15 registers, used by AVX instructions and
// YMM16-31 registers, used by AVX-512 instructions.
let SubRegIndices = [sub_xmm] in {
foreach Index = 0-31 in {
def YMM#Index : X86Reg<"ymm"#Index, Index, [!cast<X86Reg>("XMM"#Index)]>,
DwarfRegAlias<!cast<X86Reg>("XMM"#Index)>;
}
}
// ZMM Registers, used by AVX-512 instructions.
let SubRegIndices = [sub_ymm] in {
foreach Index = 0-31 in {
def ZMM#Index : X86Reg<"zmm"#Index, Index, [!cast<X86Reg>("YMM"#Index)]>,
DwarfRegAlias<!cast<X86Reg>("XMM"#Index)>;
}
}
// Tile config registers.
def TMMCFG: X86Reg<"tmmcfg", 0>;
// Tile "registers".
def TMM0: X86Reg<"tmm0", 0>;
def TMM1: X86Reg<"tmm1", 1>;
def TMM2: X86Reg<"tmm2", 2>;
def TMM3: X86Reg<"tmm3", 3>;
def TMM4: X86Reg<"tmm4", 4>;
def TMM5: X86Reg<"tmm5", 5>;
def TMM6: X86Reg<"tmm6", 6>;
def TMM7: X86Reg<"tmm7", 7>;
// Mask Registers, used by AVX-512 instructions.
def K0 : X86Reg<"k0", 0>, DwarfRegNum<[118, 93, 93]>;
def K1 : X86Reg<"k1", 1>, DwarfRegNum<[119, 94, 94]>;
def K2 : X86Reg<"k2", 2>, DwarfRegNum<[120, 95, 95]>;
def K3 : X86Reg<"k3", 3>, DwarfRegNum<[121, 96, 96]>;
def K4 : X86Reg<"k4", 4>, DwarfRegNum<[122, 97, 97]>;
def K5 : X86Reg<"k5", 5>, DwarfRegNum<[123, 98, 98]>;
def K6 : X86Reg<"k6", 6>, DwarfRegNum<[124, 99, 99]>;
def K7 : X86Reg<"k7", 7>, DwarfRegNum<[125, 100, 100]>;
// Floating point stack registers. These don't map one-to-one to the FP
// pseudo registers, but we still mark them as aliasing FP registers. That
// way both kinds can be live without exceeding the stack depth. ST registers
// are only live around inline assembly.
def ST0 : X86Reg<"st", 0>, DwarfRegNum<[33, 12, 11]>;
def ST1 : X86Reg<"st(1)", 1>, DwarfRegNum<[34, 13, 12]>;
def ST2 : X86Reg<"st(2)", 2>, DwarfRegNum<[35, 14, 13]>;
def ST3 : X86Reg<"st(3)", 3>, DwarfRegNum<[36, 15, 14]>;
def ST4 : X86Reg<"st(4)", 4>, DwarfRegNum<[37, 16, 15]>;
def ST5 : X86Reg<"st(5)", 5>, DwarfRegNum<[38, 17, 16]>;
def ST6 : X86Reg<"st(6)", 6>, DwarfRegNum<[39, 18, 17]>;
def ST7 : X86Reg<"st(7)", 7>, DwarfRegNum<[40, 19, 18]>;
// Floating-point status word
def FPSW : X86Reg<"fpsr", 0>;
// Floating-point control word
def FPCW : X86Reg<"fpcr", 0>;
// SIMD Floating-point control register.
// Note: We only model the "Uses" of the control bits: current rounding modes,
// DAZ, FTZ and exception masks. We don't model the "Defs" of flag bits.
def MXCSR : X86Reg<"mxcsr", 0>;
// Status flags register.
//
// Note that some flags that are commonly thought of as part of the status
// flags register are modeled separately. Typically this is due to instructions
// reading and updating those flags independently of all the others. We don't
// want to create false dependencies between these instructions and so we use
// a separate register to model them.
def EFLAGS : X86Reg<"flags", 0>;
// The direction flag.
def DF : X86Reg<"dirflag", 0>;
// Segment registers
def CS : X86Reg<"cs", 1>;
def DS : X86Reg<"ds", 3>;
def SS : X86Reg<"ss", 2>;
def ES : X86Reg<"es", 0>;
def FS : X86Reg<"fs", 4>;
def GS : X86Reg<"gs", 5>;
// Debug registers
def DR0 : X86Reg<"dr0", 0>;
def DR1 : X86Reg<"dr1", 1>;
def DR2 : X86Reg<"dr2", 2>;
def DR3 : X86Reg<"dr3", 3>;
def DR4 : X86Reg<"dr4", 4>;
def DR5 : X86Reg<"dr5", 5>;
def DR6 : X86Reg<"dr6", 6>;
def DR7 : X86Reg<"dr7", 7>;
def DR8 : X86Reg<"dr8", 8>;
def DR9 : X86Reg<"dr9", 9>;
def DR10 : X86Reg<"dr10", 10>;
def DR11 : X86Reg<"dr11", 11>;
def DR12 : X86Reg<"dr12", 12>;
def DR13 : X86Reg<"dr13", 13>;
def DR14 : X86Reg<"dr14", 14>;
def DR15 : X86Reg<"dr15", 15>;
// Control registers
def CR0 : X86Reg<"cr0", 0>;
def CR1 : X86Reg<"cr1", 1>;
def CR2 : X86Reg<"cr2", 2>;
def CR3 : X86Reg<"cr3", 3>;
def CR4 : X86Reg<"cr4", 4>;
def CR5 : X86Reg<"cr5", 5>;
def CR6 : X86Reg<"cr6", 6>;
def CR7 : X86Reg<"cr7", 7>;
def CR8 : X86Reg<"cr8", 8>;
def CR9 : X86Reg<"cr9", 9>;
def CR10 : X86Reg<"cr10", 10>;
def CR11 : X86Reg<"cr11", 11>;
def CR12 : X86Reg<"cr12", 12>;
def CR13 : X86Reg<"cr13", 13>;
def CR14 : X86Reg<"cr14", 14>;
def CR15 : X86Reg<"cr15", 15>;
// Pseudo index registers
def EIZ : X86Reg<"eiz", 4>;
def RIZ : X86Reg<"riz", 4>;
// Bound registers, used in MPX instructions
def BND0 : X86Reg<"bnd0", 0>;
def BND1 : X86Reg<"bnd1", 1>;
def BND2 : X86Reg<"bnd2", 2>;
def BND3 : X86Reg<"bnd3", 3>;
// CET registers - Shadow Stack Pointer
def SSP : X86Reg<"ssp", 0>;
//===----------------------------------------------------------------------===//
// Register Class Definitions... now that we have all of the pieces, define the
// top-level register classes. The order specified in the register list is
// implicitly defined to be the register allocation order.
//
// List call-clobbered registers before callee-save registers. RBX, RBP, (and
// R12, R13, R14, and R15 for X86-64) are callee-save registers.
// In 64-mode, there are 12 additional i8 registers, SIL, DIL, BPL, SPL, and
// R8B, ... R15B.
// Allocate R12 and R13 last, as these require an extra byte when
// encoded in x86_64 instructions.
// FIXME: Allow AH, CH, DH, BH to be used as general-purpose registers in
// 64-bit mode. The main complication is that they cannot be encoded in an
// instruction requiring a REX prefix, while SIL, DIL, BPL, R8D, etc.
// require a REX prefix. For example, "addb %ah, %dil" and "movzbl %ah, %r8d"
// cannot be encoded.
def GR8 : RegisterClass<"X86", [i8], 8,
(add AL, CL, DL, AH, CH, DH, BL, BH, SIL, DIL, BPL, SPL,
R8B, R9B, R10B, R11B, R14B, R15B, R12B, R13B)> {
let AltOrders = [(sub GR8, AH, BH, CH, DH)];
let AltOrderSelect = [{
return MF.getSubtarget<X86Subtarget>().is64Bit();
}];
}
let isAllocatable = 0 in
def GRH8 : RegisterClass<"X86", [i8], 8,
(add SIH, DIH, BPH, SPH, R8BH, R9BH, R10BH, R11BH,
R12BH, R13BH, R14BH, R15BH)>;
def GR16 : RegisterClass<"X86", [i16], 16,
(add AX, CX, DX, SI, DI, BX, BP, SP,
R8W, R9W, R10W, R11W, R14W, R15W, R12W, R13W)>;
let isAllocatable = 0 in
def GRH16 : RegisterClass<"X86", [i16], 16,
(add HAX, HCX, HDX, HSI, HDI, HBX, HBP, HSP, HIP,
R8WH, R9WH, R10WH, R11WH, R12WH, R13WH, R14WH,
R15WH)>;
def GR32 : RegisterClass<"X86", [i32], 32,
(add EAX, ECX, EDX, ESI, EDI, EBX, EBP, ESP,
R8D, R9D, R10D, R11D, R14D, R15D, R12D, R13D)>;
// GR64 - 64-bit GPRs. This oddly includes RIP, which isn't accurate, since
// RIP isn't really a register and it can't be used anywhere except in an
// address, but it doesn't cause trouble.
// FIXME: it *does* cause trouble - CheckBaseRegAndIndexReg() has extra
// tests because of the inclusion of RIP in this register class.
def GR64 : RegisterClass<"X86", [i64], 64,
(add RAX, RCX, RDX, RSI, RDI, R8, R9, R10, R11,
RBX, R14, R15, R12, R13, RBP, RSP, RIP)>;
// Segment registers for use by MOV instructions (and others) that have a
// segment register as one operand. Always contain a 16-bit segment
// descriptor.
def SEGMENT_REG : RegisterClass<"X86", [i16], 16, (add CS, DS, SS, ES, FS, GS)>;
// Debug registers.
def DEBUG_REG : RegisterClass<"X86", [i32], 32, (sequence "DR%u", 0, 15)>;
// Control registers.
def CONTROL_REG : RegisterClass<"X86", [i64], 64, (sequence "CR%u", 0, 15)>;
// GR8_ABCD_L, GR8_ABCD_H, GR16_ABCD, GR32_ABCD, GR64_ABCD - Subclasses of
// GR8, GR16, GR32, and GR64 which contain just the "a" "b", "c", and "d"
// registers. On x86-32, GR16_ABCD and GR32_ABCD are classes for registers
// that support 8-bit subreg operations. On x86-64, GR16_ABCD, GR32_ABCD,
// and GR64_ABCD are classes for registers that support 8-bit h-register
// operations.
def GR8_ABCD_L : RegisterClass<"X86", [i8], 8, (add AL, CL, DL, BL)>;
def GR8_ABCD_H : RegisterClass<"X86", [i8], 8, (add AH, CH, DH, BH)>;
def GR16_ABCD : RegisterClass<"X86", [i16], 16, (add AX, CX, DX, BX)>;
def GR32_ABCD : RegisterClass<"X86", [i32], 32, (add EAX, ECX, EDX, EBX)>;
def GR64_ABCD : RegisterClass<"X86", [i64], 64, (add RAX, RCX, RDX, RBX)>;
def GR32_TC : RegisterClass<"X86", [i32], 32, (add EAX, ECX, EDX, ESP)>;
def GR64_TC : RegisterClass<"X86", [i64], 64, (add RAX, RCX, RDX, RSI, RDI,
R8, R9, R11, RIP, RSP)>;
def GR64_TCW64 : RegisterClass<"X86", [i64], 64, (add RAX, RCX, RDX,
R8, R9, R10, R11,
RIP, RSP)>;
// GR8_NOREX - GR8 registers which do not require a REX prefix.
def GR8_NOREX : RegisterClass<"X86", [i8], 8,
(add AL, CL, DL, AH, CH, DH, BL, BH)> {
let AltOrders = [(sub GR8_NOREX, AH, BH, CH, DH)];
let AltOrderSelect = [{
return MF.getSubtarget<X86Subtarget>().is64Bit();
}];
}
// GR16_NOREX - GR16 registers which do not require a REX prefix.
def GR16_NOREX : RegisterClass<"X86", [i16], 16,
(add AX, CX, DX, SI, DI, BX, BP, SP)>;
// GR32_NOREX - GR32 registers which do not require a REX prefix.
def GR32_NOREX : RegisterClass<"X86", [i32], 32,
(add EAX, ECX, EDX, ESI, EDI, EBX, EBP, ESP)>;
// GR64_NOREX - GR64 registers which do not require a REX prefix.
def GR64_NOREX : RegisterClass<"X86", [i64], 64,
(add RAX, RCX, RDX, RSI, RDI, RBX, RBP, RSP, RIP)>;
// GR32_NOSP - GR32 registers except ESP.
def GR32_NOSP : RegisterClass<"X86", [i32], 32, (sub GR32, ESP)>;
// GR64_NOSP - GR64 registers except RSP (and RIP).
def GR64_NOSP : RegisterClass<"X86", [i64], 64, (sub GR64, RSP, RIP)>;
// GR32_NOREX_NOSP - GR32 registers which do not require a REX prefix except
// ESP.
def GR32_NOREX_NOSP : RegisterClass<"X86", [i32], 32,
(and GR32_NOREX, GR32_NOSP)>;
// GR64_NOREX_NOSP - GR64_NOREX registers except RSP.
def GR64_NOREX_NOSP : RegisterClass<"X86", [i64], 64,
(and GR64_NOREX, GR64_NOSP)>;
// Register classes used for ABIs that use 32-bit address accesses,
// while using the whole x84_64 ISA.
// In such cases, it is fine to use RIP as we are sure the 32 high
// bits are not set. We do not need variants for NOSP as RIP is not
// allowed there.
// RIP is not spilled anywhere for now, so stick to 32-bit alignment
// to save on memory space.
// FIXME: We could allow all 64bit registers, but we would need
// something to check that the 32 high bits are not set,
// which we do not have right now.
def LOW32_ADDR_ACCESS : RegisterClass<"X86", [i32], 32, (add GR32, RIP)>;
// When RBP is used as a base pointer in a 32-bit addresses environment,
// this is also safe to use the full register to access addresses.
// Since RBP will never be spilled, stick to a 32 alignment to save
// on memory consumption.
def LOW32_ADDR_ACCESS_RBP : RegisterClass<"X86", [i32], 32,
(add LOW32_ADDR_ACCESS, RBP)>;
// A class to support the 'A' assembler constraint: [ER]AX then [ER]DX.
def GR32_AD : RegisterClass<"X86", [i32], 32, (add EAX, EDX)>;
def GR64_AD : RegisterClass<"X86", [i64], 64, (add RAX, RDX)>;
// Classes to support the 64-bit assembler constraint tied to a fixed
// register in 32-bit mode. The second register is always the next in
// the list. Wrap around causes an error.
def GR32_DC : RegisterClass<"X86", [i32], 32, (add EDX, ECX)>;
def GR32_CB : RegisterClass<"X86", [i32], 32, (add ECX, EBX)>;
def GR32_BSI : RegisterClass<"X86", [i32], 32, (add EBX, ESI)>;
def GR32_SIDI : RegisterClass<"X86", [i32], 32, (add ESI, EDI)>;
def GR32_DIBP : RegisterClass<"X86", [i32], 32, (add EDI, EBP)>;
def GR32_BPSP : RegisterClass<"X86", [i32], 32, (add EBP, ESP)>;
// Scalar SSE2 floating point registers.
def FR32 : RegisterClass<"X86", [f32], 32, (sequence "XMM%u", 0, 15)>;
def FR64 : RegisterClass<"X86", [f64], 64, (add FR32)>;
// FIXME: This sets up the floating point register files as though they are f64
// values, though they really are f80 values. This will cause us to spill
// values as 64-bit quantities instead of 80-bit quantities, which is much much
// faster on common hardware. In reality, this should be controlled by a
// command line option or something.
def RFP32 : RegisterClass<"X86",[f32], 32, (sequence "FP%u", 0, 6)>;
def RFP64 : RegisterClass<"X86",[f64], 32, (add RFP32)>;
def RFP80 : RegisterClass<"X86",[f80], 32, (add RFP32)>;
// st(7) may be is not allocatable.
def RFP80_7 : RegisterClass<"X86",[f80], 32, (add FP7)> {
let isAllocatable = 0;
}
// Floating point stack registers (these are not allocatable by the
// register allocator - the floating point stackifier is responsible
// for transforming FPn allocations to STn registers)
def RST : RegisterClass<"X86", [f80, f64, f32], 32, (sequence "ST%u", 0, 7)> {
let isAllocatable = 0;
}
// Helper to allow %st to print as %st(0) when its encoded in the instruction.
def RSTi : RegisterOperand<RST, "printSTiRegOperand">;
// Generic vector registers: VR64 and VR128.
// Ensure that float types are declared first - only float is legal on SSE1.
def VR64: RegisterClass<"X86", [x86mmx], 64, (sequence "MM%u", 0, 7)>;
def VR128 : RegisterClass<"X86", [v4f32, v2f64, v16i8, v8i16, v4i32, v2i64, f128],
128, (add FR32)>;
def VR256 : RegisterClass<"X86", [v8f32, v4f64, v32i8, v16i16, v8i32, v4i64],
256, (sequence "YMM%u", 0, 15)>;
// Status flags registers.
def CCR : RegisterClass<"X86", [i32], 32, (add EFLAGS)> {
let CopyCost = -1; // Don't allow copying of status registers.
let isAllocatable = 0;
}
def FPCCR : RegisterClass<"X86", [i16], 16, (add FPSW)> {
let CopyCost = -1; // Don't allow copying of status registers.
let isAllocatable = 0;
}
def DFCCR : RegisterClass<"X86", [i32], 32, (add DF)> {
let CopyCost = -1; // Don't allow copying of status registers.
let isAllocatable = 0;
}
// AVX-512 vector/mask registers.
def VR512 : RegisterClass<"X86", [v16f32, v8f64, v64i8, v32i16, v16i32, v8i64],
512, (sequence "ZMM%u", 0, 31)>;
// Represents the lower 16 registers that have VEX/legacy encodable subregs.
def VR512_0_15 : RegisterClass<"X86", [v16f32, v8f64, v64i8, v32i16, v16i32, v8i64],
512, (sequence "ZMM%u", 0, 15)>;
// Scalar AVX-512 floating point registers.
def FR32X : RegisterClass<"X86", [f32], 32, (sequence "XMM%u", 0, 31)>;
def FR64X : RegisterClass<"X86", [f64], 64, (add FR32X)>;
// Extended VR128 and VR256 for AVX-512 instructions
def VR128X : RegisterClass<"X86", [v4f32, v2f64, v16i8, v8i16, v4i32, v2i64, f128],
128, (add FR32X)>;
def VR256X : RegisterClass<"X86", [v8f32, v4f64, v32i8, v16i16, v8i32, v4i64],
256, (sequence "YMM%u", 0, 31)>;
// Mask registers
def VK1 : RegisterClass<"X86", [v1i1], 16, (sequence "K%u", 0, 7)> {let Size = 16;}
def VK2 : RegisterClass<"X86", [v2i1], 16, (add VK1)> {let Size = 16;}
def VK4 : RegisterClass<"X86", [v4i1], 16, (add VK2)> {let Size = 16;}
def VK8 : RegisterClass<"X86", [v8i1], 16, (add VK4)> {let Size = 16;}
def VK16 : RegisterClass<"X86", [v16i1], 16, (add VK8)> {let Size = 16;}
def VK32 : RegisterClass<"X86", [v32i1], 32, (add VK16)> {let Size = 32;}
def VK64 : RegisterClass<"X86", [v64i1], 64, (add VK32)> {let Size = 64;}
// Mask register pairs
def KPAIRS : RegisterTuples<[sub_mask_0, sub_mask_1],
[(add K0, K2, K4, K6), (add K1, K3, K5, K7)]>;
def VK1PAIR : RegisterClass<"X86", [untyped], 16, (add KPAIRS)> {let Size = 32;}
def VK2PAIR : RegisterClass<"X86", [untyped], 16, (add KPAIRS)> {let Size = 32;}
def VK4PAIR : RegisterClass<"X86", [untyped], 16, (add KPAIRS)> {let Size = 32;}
def VK8PAIR : RegisterClass<"X86", [untyped], 16, (add KPAIRS)> {let Size = 32;}
def VK16PAIR : RegisterClass<"X86", [untyped], 16, (add KPAIRS)> {let Size = 32;}
def VK1WM : RegisterClass<"X86", [v1i1], 16, (sub VK1, K0)> {let Size = 16;}
def VK2WM : RegisterClass<"X86", [v2i1], 16, (sub VK2, K0)> {let Size = 16;}
def VK4WM : RegisterClass<"X86", [v4i1], 16, (sub VK4, K0)> {let Size = 16;}
def VK8WM : RegisterClass<"X86", [v8i1], 16, (sub VK8, K0)> {let Size = 16;}
def VK16WM : RegisterClass<"X86", [v16i1], 16, (add VK8WM)> {let Size = 16;}
def VK32WM : RegisterClass<"X86", [v32i1], 32, (add VK16WM)> {let Size = 32;}
def VK64WM : RegisterClass<"X86", [v64i1], 64, (add VK32WM)> {let Size = 64;}
// Bound registers
def BNDR : RegisterClass<"X86", [v2i64], 128, (sequence "BND%u", 0, 3)>;
// Tiles
let CopyCost = -1 in // Don't allow copying of tile registers
def TILE : RegisterClass<"X86", [x86amx], 8192,
(sequence "TMM%u", 0, 7)> {let Size = 8192;}
def TILECFG : RegisterClass<"X86", [untyped], 512, (add TMMCFG)> {
let CopyCost = -1; // Don't allow copying of tile config registers.
let isAllocatable = 1;
let Size = 512;
}
|