llvm.org GIT mirror llvm / release_50 lib / Target / AMDGPU / VOP3Instructions.td
release_50

Tree @release_50 (Download .tar.gz)

VOP3Instructions.td @release_50raw · history · blame

  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
//===-- VOP3Instructions.td - Vector Instruction Defintions ---------------===//
//
//                     The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//

//===----------------------------------------------------------------------===//
// VOP3 Classes
//===----------------------------------------------------------------------===//

class getVOP3ModPat<VOPProfile P, SDPatternOperator node> {
  dag src0 = !if(P.HasOMod,
    (VOP3Mods0 P.Src0VT:$src0, i32:$src0_modifiers, i1:$clamp, i32:$omod),
    (VOP3Mods0 P.Src0VT:$src0, i32:$src0_modifiers, i1:$clamp));

  list<dag> ret3 = [(set P.DstVT:$vdst,
    (node (P.Src0VT src0),
          (P.Src1VT (VOP3Mods P.Src1VT:$src1, i32:$src1_modifiers)),
          (P.Src2VT (VOP3Mods P.Src2VT:$src2, i32:$src2_modifiers))))];

  list<dag> ret2 = [(set P.DstVT:$vdst,
    (node (P.Src0VT src0),
          (P.Src1VT (VOP3Mods P.Src1VT:$src1, i32:$src1_modifiers))))];

  list<dag> ret1 = [(set P.DstVT:$vdst,
    (node (P.Src0VT src0)))];

  list<dag> ret = !if(!eq(P.NumSrcArgs, 3), ret3,
                  !if(!eq(P.NumSrcArgs, 2), ret2,
                  ret1));
}

class getVOP3PModPat<VOPProfile P, SDPatternOperator node> {
  list<dag> ret3 = [(set P.DstVT:$vdst,
    (node (P.Src0VT !if(P.HasClamp, (VOP3PMods0 P.Src0VT:$src0, i32:$src0_modifiers, i1:$clamp),
                                    (VOP3PMods P.Src0VT:$src0, i32:$src0_modifiers))),
          (P.Src1VT (VOP3PMods P.Src1VT:$src1, i32:$src1_modifiers)),
          (P.Src2VT (VOP3PMods P.Src2VT:$src2, i32:$src2_modifiers))))];

  list<dag> ret2 = [(set P.DstVT:$vdst,
    (node !if(P.HasClamp, (P.Src0VT (VOP3PMods0 P.Src0VT:$src0, i32:$src0_modifiers, i1:$clamp)),
                          (P.Src0VT (VOP3PMods P.Src0VT:$src0, i32:$src0_modifiers))),
          (P.Src1VT (VOP3PMods P.Src1VT:$src1, i32:$src1_modifiers))))];

  list<dag> ret1 = [(set P.DstVT:$vdst,
    (node (P.Src0VT (VOP3PMods0 P.Src0VT:$src0, i32:$src0_modifiers, i1:$clamp))))];

  list<dag> ret = !if(!eq(P.NumSrcArgs, 3), ret3,
                  !if(!eq(P.NumSrcArgs, 2), ret2,
                  ret1));
}

class getVOP3Pat<VOPProfile P, SDPatternOperator node> {
  list<dag> ret3 = [(set P.DstVT:$vdst, (node P.Src0VT:$src0, P.Src1VT:$src1, P.Src2VT:$src2))];
  list<dag> ret2 = [(set P.DstVT:$vdst, (node P.Src0VT:$src0, P.Src1VT:$src1))];
  list<dag> ret1 = [(set P.DstVT:$vdst, (node P.Src0VT:$src0))];
  list<dag> ret = !if(!eq(P.NumSrcArgs, 3), ret3,
                  !if(!eq(P.NumSrcArgs, 2), ret2,
                  ret1));
}

class VOP3Inst<string OpName, VOPProfile P, SDPatternOperator node = null_frag, bit VOP3Only = 0> :
  VOP3_Pseudo<OpName, P,
    !if(P.HasModifiers, getVOP3ModPat<P, node>.ret, getVOP3Pat<P, node>.ret),
    VOP3Only>;

// Special case for v_div_fmas_{f32|f64}, since it seems to be the
// only VOP instruction that implicitly reads VCC.
let Asm64 = " $vdst, $src0_modifiers, $src1_modifiers, $src2_modifiers$clamp$omod" in {
def VOP_F32_F32_F32_F32_VCC : VOPProfile<[f32, f32, f32, f32]> {
  let Outs64 = (outs DstRC.RegClass:$vdst);
}
def VOP_F64_F64_F64_F64_VCC : VOPProfile<[f64, f64, f64, f64]> {
  let Outs64 = (outs DstRC.RegClass:$vdst);
}
}

class getVOP3VCC<VOPProfile P, SDPatternOperator node> {
  list<dag> ret =
    [(set P.DstVT:$vdst,
      (node (P.Src0VT (VOP3Mods0 P.Src0VT:$src0, i32:$src0_modifiers, i1:$clamp, i32:$omod)),
            (P.Src1VT (VOP3Mods P.Src1VT:$src1, i32:$src1_modifiers)),
            (P.Src2VT (VOP3Mods P.Src2VT:$src2, i32:$src2_modifiers)),
            (i1 VCC)))];
}

class VOP3_Profile<VOPProfile P> : VOPProfile<P.ArgVT> {
  // FIXME: Hack to stop printing _e64
  let Outs64 = (outs DstRC.RegClass:$vdst);
  let Asm64 = " " # P.Asm64;
}

class VOP3b_Profile<ValueType vt> : VOPProfile<[vt, vt, vt, vt]> {
  // v_div_scale_{f32|f64} do not support input modifiers.
  let HasModifiers = 0;
  let HasOMod = 0;
  let Outs64 = (outs DstRC:$vdst, SReg_64:$sdst);
  let Asm64 = " $vdst, $sdst, $src0, $src1, $src2";
}

def VOP3b_F32_I1_F32_F32_F32 : VOP3b_Profile<f32> {
  // FIXME: Hack to stop printing _e64
  let DstRC = RegisterOperand<VGPR_32>;
}

def VOP3b_F64_I1_F64_F64_F64 : VOP3b_Profile<f64> {
  // FIXME: Hack to stop printing _e64
  let DstRC = RegisterOperand<VReg_64>;
}

def VOP3b_I64_I1_I32_I32_I64 : VOPProfile<[i64, i32, i32, i64]> {
  // FIXME: Hack to stop printing _e64
  let DstRC = RegisterOperand<VReg_64>;

  let Outs64 = (outs DstRC:$vdst, SReg_64:$sdst);
  let Asm64 = " $vdst, $sdst, $src0, $src1, $src2";
}

//===----------------------------------------------------------------------===//
// VOP3 Instructions
//===----------------------------------------------------------------------===//

let isCommutable = 1 in {

def V_MAD_LEGACY_F32 : VOP3Inst <"v_mad_legacy_f32", VOP3_Profile<VOP_F32_F32_F32_F32>>;
def V_MAD_F32 : VOP3Inst <"v_mad_f32", VOP3_Profile<VOP_F32_F32_F32_F32>, fmad>;
def V_MAD_I32_I24 : VOP3Inst <"v_mad_i32_i24", VOP3_Profile<VOP_I32_I32_I32_I32>, AMDGPUmad_i24>;
def V_MAD_U32_U24 : VOP3Inst <"v_mad_u32_u24", VOP3_Profile<VOP_I32_I32_I32_I32>, AMDGPUmad_u24>;
def V_FMA_F32 : VOP3Inst <"v_fma_f32", VOP3_Profile<VOP_F32_F32_F32_F32>, fma>;
def V_FMA_F64 : VOP3Inst <"v_fma_f64", VOP3_Profile<VOP_F64_F64_F64_F64>, fma>;
def V_LERP_U8 : VOP3Inst <"v_lerp_u8", VOP3_Profile<VOP_I32_I32_I32_I32>, int_amdgcn_lerp>;

let SchedRW = [WriteDoubleAdd] in {
def V_ADD_F64 : VOP3Inst <"v_add_f64", VOP3_Profile<VOP_F64_F64_F64>, fadd, 1>;
def V_MUL_F64 : VOP3Inst <"v_mul_f64", VOP3_Profile<VOP_F64_F64_F64>, fmul, 1>;
def V_MIN_F64 : VOP3Inst <"v_min_f64", VOP3_Profile<VOP_F64_F64_F64>, fminnum, 1>;
def V_MAX_F64 : VOP3Inst <"v_max_f64", VOP3_Profile<VOP_F64_F64_F64>, fmaxnum, 1>;
} // End SchedRW = [WriteDoubleAdd]

let SchedRW = [WriteQuarterRate32] in {
def V_MUL_LO_U32 : VOP3Inst <"v_mul_lo_u32", VOP3_Profile<VOP_I32_I32_I32>>;
def V_MUL_HI_U32 : VOP3Inst <"v_mul_hi_u32", VOP3_Profile<VOP_I32_I32_I32>, mulhu>;
def V_MUL_LO_I32 : VOP3Inst <"v_mul_lo_i32", VOP3_Profile<VOP_I32_I32_I32>>;
def V_MUL_HI_I32 : VOP3Inst <"v_mul_hi_i32", VOP3_Profile<VOP_I32_I32_I32>, mulhs>;
} // End SchedRW = [WriteQuarterRate32]

let Uses = [VCC, EXEC] in {
// v_div_fmas_f32:
//   result = src0 * src1 + src2
//   if (vcc)
//     result *= 2^32
//
def V_DIV_FMAS_F32 : VOP3_Pseudo <"v_div_fmas_f32", VOP_F32_F32_F32_F32_VCC,
  getVOP3VCC<VOP_F32_F32_F32_F32_VCC, AMDGPUdiv_fmas>.ret> {
  let SchedRW = [WriteFloatFMA];
}
// v_div_fmas_f64:
//   result = src0 * src1 + src2
//   if (vcc)
//     result *= 2^64
//
def V_DIV_FMAS_F64 : VOP3_Pseudo <"v_div_fmas_f64", VOP_F64_F64_F64_F64_VCC,
  getVOP3VCC<VOP_F64_F64_F64_F64_VCC, AMDGPUdiv_fmas>.ret> {
  let SchedRW = [WriteDouble];
}
} // End Uses = [VCC, EXEC]

} // End isCommutable = 1

def V_CUBEID_F32 : VOP3Inst <"v_cubeid_f32", VOP3_Profile<VOP_F32_F32_F32_F32>, int_amdgcn_cubeid>;
def V_CUBESC_F32 : VOP3Inst <"v_cubesc_f32", VOP3_Profile<VOP_F32_F32_F32_F32>, int_amdgcn_cubesc>;
def V_CUBETC_F32 : VOP3Inst <"v_cubetc_f32", VOP3_Profile<VOP_F32_F32_F32_F32>, int_amdgcn_cubetc>;
def V_CUBEMA_F32 : VOP3Inst <"v_cubema_f32", VOP3_Profile<VOP_F32_F32_F32_F32>, int_amdgcn_cubema>;
def V_BFE_U32 : VOP3Inst <"v_bfe_u32", VOP3_Profile<VOP_I32_I32_I32_I32>, AMDGPUbfe_u32>;
def V_BFE_I32 : VOP3Inst <"v_bfe_i32", VOP3_Profile<VOP_I32_I32_I32_I32>, AMDGPUbfe_i32>;
def V_BFI_B32 : VOP3Inst <"v_bfi_b32", VOP3_Profile<VOP_I32_I32_I32_I32>, AMDGPUbfi>;
def V_ALIGNBIT_B32 : VOP3Inst <"v_alignbit_b32", VOP3_Profile<VOP_I32_I32_I32_I32>, int_amdgcn_alignbit>;
def V_ALIGNBYTE_B32 : VOP3Inst <"v_alignbyte_b32", VOP3_Profile<VOP_I32_I32_I32_I32>, int_amdgcn_alignbyte>;
def V_MIN3_F32 : VOP3Inst <"v_min3_f32", VOP3_Profile<VOP_F32_F32_F32_F32>, AMDGPUfmin3>;
def V_MIN3_I32 : VOP3Inst <"v_min3_i32", VOP3_Profile<VOP_I32_I32_I32_I32>, AMDGPUsmin3>;
def V_MIN3_U32 : VOP3Inst <"v_min3_u32", VOP3_Profile<VOP_I32_I32_I32_I32>, AMDGPUumin3>;
def V_MAX3_F32 : VOP3Inst <"v_max3_f32", VOP3_Profile<VOP_F32_F32_F32_F32>, AMDGPUfmax3>;
def V_MAX3_I32 : VOP3Inst <"v_max3_i32", VOP3_Profile<VOP_I32_I32_I32_I32>, AMDGPUsmax3>;
def V_MAX3_U32 : VOP3Inst <"v_max3_u32", VOP3_Profile<VOP_I32_I32_I32_I32>, AMDGPUumax3>;
def V_MED3_F32 : VOP3Inst <"v_med3_f32", VOP3_Profile<VOP_F32_F32_F32_F32>, AMDGPUfmed3>;
def V_MED3_I32 : VOP3Inst <"v_med3_i32", VOP3_Profile<VOP_I32_I32_I32_I32>, AMDGPUsmed3>;
def V_MED3_U32 : VOP3Inst <"v_med3_u32", VOP3_Profile<VOP_I32_I32_I32_I32>, AMDGPUumed3>;
def V_SAD_U8 : VOP3Inst <"v_sad_u8", VOP3_Profile<VOP_I32_I32_I32_I32>, int_amdgcn_sad_u8>;
def V_SAD_HI_U8 : VOP3Inst <"v_sad_hi_u8", VOP3_Profile<VOP_I32_I32_I32_I32>, int_amdgcn_sad_hi_u8>;
def V_SAD_U16 : VOP3Inst <"v_sad_u16", VOP3_Profile<VOP_I32_I32_I32_I32>, int_amdgcn_sad_u16>;
def V_SAD_U32 : VOP3Inst <"v_sad_u32", VOP3_Profile<VOP_I32_I32_I32_I32>>;
def V_CVT_PK_U8_F32 : VOP3Inst<"v_cvt_pk_u8_f32", VOP3_Profile<VOP_I32_F32_I32_I32>, int_amdgcn_cvt_pk_u8_f32>;
def V_DIV_FIXUP_F32 : VOP3Inst <"v_div_fixup_f32", VOP3_Profile<VOP_F32_F32_F32_F32>, AMDGPUdiv_fixup>;

let SchedRW = [WriteDoubleAdd] in {
def V_DIV_FIXUP_F64 : VOP3Inst <"v_div_fixup_f64", VOP3_Profile<VOP_F64_F64_F64_F64>, AMDGPUdiv_fixup>;
def V_LDEXP_F64 : VOP3Inst <"v_ldexp_f64", VOP3_Profile<VOP_F64_F64_I32>, AMDGPUldexp, 1>;
} // End SchedRW = [WriteDoubleAdd]

def V_DIV_SCALE_F32 : VOP3_Pseudo <"v_div_scale_f32", VOP3b_F32_I1_F32_F32_F32, [], 1> {
  let SchedRW = [WriteFloatFMA, WriteSALU];
  let hasExtraSrcRegAllocReq = 1;
  let AsmMatchConverter = "";
}

// Double precision division pre-scale.
def V_DIV_SCALE_F64 : VOP3_Pseudo <"v_div_scale_f64", VOP3b_F64_I1_F64_F64_F64, [], 1> {
  let SchedRW = [WriteDouble, WriteSALU];
  let hasExtraSrcRegAllocReq = 1;
  let AsmMatchConverter = "";
}

def V_MSAD_U8 : VOP3Inst <"v_msad_u8", VOP3_Profile<VOP_I32_I32_I32_I32>, int_amdgcn_msad_u8>;

let Constraints = "@earlyclobber $vdst" in {
def V_MQSAD_PK_U16_U8 : VOP3Inst <"v_mqsad_pk_u16_u8", VOP3_Profile<VOP_I64_I64_I32_I64>, int_amdgcn_mqsad_pk_u16_u8>;
} // End Constraints = "@earlyclobber $vdst"

def V_TRIG_PREOP_F64 : VOP3Inst <"v_trig_preop_f64", VOP3_Profile<VOP_F64_F64_I32>, AMDGPUtrig_preop> {
  let SchedRW = [WriteDouble];
}

// These instructions only exist on SI and CI
let SubtargetPredicate = isSICI in {
def V_LSHL_B64 : VOP3Inst <"v_lshl_b64", VOP3_Profile<VOP_I64_I64_I32>>;
def V_LSHR_B64 : VOP3Inst <"v_lshr_b64", VOP3_Profile<VOP_I64_I64_I32>>;
def V_ASHR_I64 : VOP3Inst <"v_ashr_i64", VOP3_Profile<VOP_I64_I64_I32>>;
def V_MULLIT_F32 : VOP3Inst <"v_mullit_f32", VOP3_Profile<VOP_F32_F32_F32_F32>>;
} // End SubtargetPredicate = isSICI

let SubtargetPredicate = isVI in {
def V_LSHLREV_B64 : VOP3Inst <"v_lshlrev_b64", VOP3_Profile<VOP_I64_I32_I64>>;
def V_LSHRREV_B64 : VOP3Inst <"v_lshrrev_b64", VOP3_Profile<VOP_I64_I32_I64>>;
def V_ASHRREV_I64 : VOP3Inst <"v_ashrrev_i64", VOP3_Profile<VOP_I64_I32_I64>>;
} // End SubtargetPredicate = isVI


let SubtargetPredicate = isCIVI in {

let Constraints = "@earlyclobber $vdst" in {
def V_QSAD_PK_U16_U8 : VOP3Inst <"v_qsad_pk_u16_u8", VOP3_Profile<VOP_I64_I64_I32_I64>, int_amdgcn_qsad_pk_u16_u8>;
def V_MQSAD_U32_U8 : VOP3Inst <"v_mqsad_u32_u8", VOP3_Profile<VOP_V4I32_I64_I32_V4I32>, int_amdgcn_mqsad_u32_u8>;
} // End Constraints = "@earlyclobber $vdst"

let isCommutable = 1 in {
def V_MAD_U64_U32 : VOP3Inst <"v_mad_u64_u32", VOP3b_I64_I1_I32_I32_I64>;
def V_MAD_I64_I32 : VOP3Inst <"v_mad_i64_i32", VOP3b_I64_I1_I32_I32_I64>;
} // End isCommutable = 1

} // End SubtargetPredicate = isCIVI


let SubtargetPredicate = Has16BitInsts in {

def V_DIV_FIXUP_F16   : VOP3Inst <"v_div_fixup_f16", VOP3_Profile<VOP_F16_F16_F16_F16>, AMDGPUdiv_fixup>;

let isCommutable = 1 in {

def V_FMA_F16         : VOP3Inst <"v_fma_f16", VOP3_Profile<VOP_F16_F16_F16_F16>, fma>;
def V_INTERP_P1LL_F16 : VOP3Inst <"v_interp_p1ll_f16", VOP3_Profile<VOP_F32_F32_F16>>;
def V_INTERP_P1LV_F16 : VOP3Inst <"v_interp_p1lv_f16", VOP3_Profile<VOP_F32_F32_F16_F16>>;
def V_INTERP_P2_F16   : VOP3Inst <"v_interp_p2_f16", VOP3_Profile<VOP_F16_F32_F16_F32>>;
def V_MAD_F16         : VOP3Inst <"v_mad_f16", VOP3_Profile<VOP_F16_F16_F16_F16>, fmad>;

def V_MAD_U16 : VOP3Inst <"v_mad_u16", VOP3_Profile<VOP_I16_I16_I16_I16>>;
def V_MAD_I16 : VOP3Inst <"v_mad_i16", VOP3_Profile<VOP_I16_I16_I16_I16>>;

}  // End isCommutable = 1
} // End SubtargetPredicate = Has16BitInsts

let SubtargetPredicate = isVI in {
def V_PERM_B32 : VOP3Inst <"v_perm_b32", VOP3_Profile<VOP_I32_I32_I32_I32>>;
} // End SubtargetPredicate = isVI

let Predicates = [Has16BitInsts] in {

multiclass Ternary_i16_Pats <SDPatternOperator op1, SDPatternOperator op2,
                             Instruction inst, SDPatternOperator op3> {
def : Pat<
  (op2 (op1 i16:$src0, i16:$src1), i16:$src2),
  (inst i16:$src0, i16:$src1, i16:$src2)
>;

def : Pat<
  (i32 (op3 (op2 (op1 i16:$src0, i16:$src1), i16:$src2))),
  (inst i16:$src0, i16:$src1, i16:$src2)
>;

def : Pat<
  (i64 (op3 (op2 (op1 i16:$src0, i16:$src1), i16:$src2))),
   (REG_SEQUENCE VReg_64,
     (inst i16:$src0, i16:$src1, i16:$src2), sub0,
     (V_MOV_B32_e32 (i32 0)), sub1)
>;
}

defm: Ternary_i16_Pats<mul, add, V_MAD_U16, zext>;
defm: Ternary_i16_Pats<mul, add, V_MAD_I16, sext>;

} // End Predicates = [Has16BitInsts]

let SubtargetPredicate = isGFX9 in {
def V_PACK_B32_F16 : VOP3Inst <"v_pack_b32_f16", VOP3_Profile<VOP_B32_F16_F16>>;
def V_LSHL_ADD_U32 : VOP3Inst <"v_lshl_add_u32", VOP3_Profile<VOP_I32_I32_I32_I32>>;
def V_ADD_LSHL_U32 : VOP3Inst <"v_add_lshl_u32", VOP3_Profile<VOP_I32_I32_I32_I32>>;
def V_ADD3_U32 : VOP3Inst <"v_add3_u32", VOP3_Profile<VOP_I32_I32_I32_I32>>;
def V_LSHL_OR_B32 : VOP3Inst <"v_lshl_or_b32", VOP3_Profile<VOP_I32_I32_I32_I32>>;
def V_AND_OR_B32 : VOP3Inst <"v_and_or_b32", VOP3_Profile<VOP_I32_I32_I32_I32>>;
def V_OR3_B32 : VOP3Inst <"v_or3_b32", VOP3_Profile<VOP_I32_I32_I32_I32>>;

def V_XAD_U32 : VOP3Inst <"v_xad_u32", VOP3_Profile<VOP_I32_I32_I32_I32>>;

def V_MED3_F16 : VOP3Inst <"v_med3_f16", VOP3_Profile<VOP_F16_F16_F16_F16>, AMDGPUfmed3>;
def V_MED3_I16 : VOP3Inst <"v_med3_i16", VOP3_Profile<VOP_I16_I16_I16_I16>, AMDGPUsmed3>;
def V_MED3_U16 : VOP3Inst <"v_med3_u16", VOP3_Profile<VOP_I16_I16_I16_I16>, AMDGPUumed3>;

def V_MIN3_F16 : VOP3Inst <"v_min3_f16", VOP3_Profile<VOP_F16_F16_F16_F16>, AMDGPUfmin3>;
def V_MIN3_I16 : VOP3Inst <"v_min3_i16", VOP3_Profile<VOP_I16_I16_I16_I16>, AMDGPUsmin3>;
def V_MIN3_U16 : VOP3Inst <"v_min3_u16", VOP3_Profile<VOP_I16_I16_I16_I16>, AMDGPUumin3>;

def V_MAX3_F16 : VOP3Inst <"v_max3_f16", VOP3_Profile<VOP_F16_F16_F16_F16>, AMDGPUfmax3>;
def V_MAX3_I16 : VOP3Inst <"v_max3_i16", VOP3_Profile<VOP_I16_I16_I16_I16>, AMDGPUsmax3>;
def V_MAX3_U16 : VOP3Inst <"v_max3_u16", VOP3_Profile<VOP_I16_I16_I16_I16>, AMDGPUumax3>;
} // End SubtargetPredicate = isGFX9


//===----------------------------------------------------------------------===//
// Target
//===----------------------------------------------------------------------===//

//===----------------------------------------------------------------------===//
// SI
//===----------------------------------------------------------------------===//

let AssemblerPredicates = [isSICI], DecoderNamespace = "SICI" in {

multiclass VOP3_Real_si<bits<9> op> {
  def _si : VOP3_Real<!cast<VOP3_Pseudo>(NAME), SIEncodingFamily.SI>,
            VOP3e_si <op, !cast<VOP3_Pseudo>(NAME).Pfl>;
}

multiclass VOP3be_Real_si<bits<9> op> {
  def _si : VOP3_Real<!cast<VOP3_Pseudo>(NAME), SIEncodingFamily.SI>,
            VOP3be_si <op, !cast<VOP3_Pseudo>(NAME).Pfl>;
}

} // End AssemblerPredicates = [isSICI], DecoderNamespace = "SICI"

defm V_MAD_LEGACY_F32   : VOP3_Real_si <0x140>;
defm V_MAD_F32          : VOP3_Real_si <0x141>;
defm V_MAD_I32_I24      : VOP3_Real_si <0x142>;
defm V_MAD_U32_U24      : VOP3_Real_si <0x143>;
defm V_CUBEID_F32       : VOP3_Real_si <0x144>;
defm V_CUBESC_F32       : VOP3_Real_si <0x145>;
defm V_CUBETC_F32       : VOP3_Real_si <0x146>;
defm V_CUBEMA_F32       : VOP3_Real_si <0x147>;
defm V_BFE_U32          : VOP3_Real_si <0x148>;
defm V_BFE_I32          : VOP3_Real_si <0x149>;
defm V_BFI_B32          : VOP3_Real_si <0x14a>;
defm V_FMA_F32          : VOP3_Real_si <0x14b>;
defm V_FMA_F64          : VOP3_Real_si <0x14c>;
defm V_LERP_U8          : VOP3_Real_si <0x14d>;
defm V_ALIGNBIT_B32     : VOP3_Real_si <0x14e>;
defm V_ALIGNBYTE_B32    : VOP3_Real_si <0x14f>;
defm V_MULLIT_F32       : VOP3_Real_si <0x150>;
defm V_MIN3_F32         : VOP3_Real_si <0x151>;
defm V_MIN3_I32         : VOP3_Real_si <0x152>;
defm V_MIN3_U32         : VOP3_Real_si <0x153>;
defm V_MAX3_F32         : VOP3_Real_si <0x154>;
defm V_MAX3_I32         : VOP3_Real_si <0x155>;
defm V_MAX3_U32         : VOP3_Real_si <0x156>;
defm V_MED3_F32         : VOP3_Real_si <0x157>;
defm V_MED3_I32         : VOP3_Real_si <0x158>;
defm V_MED3_U32         : VOP3_Real_si <0x159>;
defm V_SAD_U8           : VOP3_Real_si <0x15a>;
defm V_SAD_HI_U8        : VOP3_Real_si <0x15b>;
defm V_SAD_U16          : VOP3_Real_si <0x15c>;
defm V_SAD_U32          : VOP3_Real_si <0x15d>;
defm V_CVT_PK_U8_F32    : VOP3_Real_si <0x15e>;
defm V_DIV_FIXUP_F32    : VOP3_Real_si <0x15f>;
defm V_DIV_FIXUP_F64    : VOP3_Real_si <0x160>;
defm V_LSHL_B64         : VOP3_Real_si <0x161>;
defm V_LSHR_B64         : VOP3_Real_si <0x162>;
defm V_ASHR_I64         : VOP3_Real_si <0x163>;
defm V_ADD_F64          : VOP3_Real_si <0x164>;
defm V_MUL_F64          : VOP3_Real_si <0x165>;
defm V_MIN_F64          : VOP3_Real_si <0x166>;
defm V_MAX_F64          : VOP3_Real_si <0x167>;
defm V_LDEXP_F64        : VOP3_Real_si <0x168>;
defm V_MUL_LO_U32       : VOP3_Real_si <0x169>;
defm V_MUL_HI_U32       : VOP3_Real_si <0x16a>;
defm V_MUL_LO_I32       : VOP3_Real_si <0x16b>;
defm V_MUL_HI_I32       : VOP3_Real_si <0x16c>;
defm V_DIV_SCALE_F32    : VOP3be_Real_si <0x16d>;
defm V_DIV_SCALE_F64    : VOP3be_Real_si <0x16e>;
defm V_DIV_FMAS_F32     : VOP3_Real_si <0x16f>;
defm V_DIV_FMAS_F64     : VOP3_Real_si <0x170>;
defm V_MSAD_U8          : VOP3_Real_si <0x171>;
defm V_MQSAD_PK_U16_U8  : VOP3_Real_si <0x173>;
defm V_TRIG_PREOP_F64   : VOP3_Real_si <0x174>;

//===----------------------------------------------------------------------===//
// CI
//===----------------------------------------------------------------------===//

multiclass VOP3_Real_ci<bits<9> op> {
  def _ci : VOP3_Real<!cast<VOP3_Pseudo>(NAME), SIEncodingFamily.SI>,
            VOP3e_si <op, !cast<VOP3_Pseudo>(NAME).Pfl> {
    let AssemblerPredicates = [isCIOnly];
    let DecoderNamespace = "CI";
  }
}

multiclass VOP3be_Real_ci<bits<9> op> {
  def _ci : VOP3_Real<!cast<VOP3_Pseudo>(NAME), SIEncodingFamily.SI>,
            VOP3be_si <op, !cast<VOP3_Pseudo>(NAME).Pfl> {
    let AssemblerPredicates = [isCIOnly];
    let DecoderNamespace = "CI";
  }
}

defm V_QSAD_PK_U16_U8   : VOP3_Real_ci <0x172>;
defm V_MQSAD_U32_U8     : VOP3_Real_ci <0x175>;
defm V_MAD_U64_U32      : VOP3be_Real_ci <0x176>;
defm V_MAD_I64_I32      : VOP3be_Real_ci <0x177>;

//===----------------------------------------------------------------------===//
// VI
//===----------------------------------------------------------------------===//

let AssemblerPredicates = [isVI], DecoderNamespace = "VI" in {

multiclass VOP3_Real_vi<bits<10> op> {
  def _vi : VOP3_Real<!cast<VOP3_Pseudo>(NAME), SIEncodingFamily.VI>,
            VOP3e_vi <op, !cast<VOP3_Pseudo>(NAME).Pfl>;
}

multiclass VOP3be_Real_vi<bits<10> op> {
  def _vi : VOP3_Real<!cast<VOP3_Pseudo>(NAME), SIEncodingFamily.VI>,
            VOP3be_vi <op, !cast<VOP3_Pseudo>(NAME).Pfl>;
}

} // End AssemblerPredicates = [isVI], DecoderNamespace = "VI"

defm V_MAD_U64_U32      : VOP3be_Real_vi <0x1E8>;
defm V_MAD_I64_I32      : VOP3be_Real_vi <0x1E9>;

defm V_MAD_LEGACY_F32   : VOP3_Real_vi <0x1c0>;
defm V_MAD_F32          : VOP3_Real_vi <0x1c1>;
defm V_MAD_I32_I24      : VOP3_Real_vi <0x1c2>;
defm V_MAD_U32_U24      : VOP3_Real_vi <0x1c3>;
defm V_CUBEID_F32       : VOP3_Real_vi <0x1c4>;
defm V_CUBESC_F32       : VOP3_Real_vi <0x1c5>;
defm V_CUBETC_F32       : VOP3_Real_vi <0x1c6>;
defm V_CUBEMA_F32       : VOP3_Real_vi <0x1c7>;
defm V_BFE_U32          : VOP3_Real_vi <0x1c8>;
defm V_BFE_I32          : VOP3_Real_vi <0x1c9>;
defm V_BFI_B32          : VOP3_Real_vi <0x1ca>;
defm V_FMA_F32          : VOP3_Real_vi <0x1cb>;
defm V_FMA_F64          : VOP3_Real_vi <0x1cc>;
defm V_LERP_U8          : VOP3_Real_vi <0x1cd>;
defm V_ALIGNBIT_B32     : VOP3_Real_vi <0x1ce>;
defm V_ALIGNBYTE_B32    : VOP3_Real_vi <0x1cf>;
defm V_MIN3_F32         : VOP3_Real_vi <0x1d0>;
defm V_MIN3_I32         : VOP3_Real_vi <0x1d1>;
defm V_MIN3_U32         : VOP3_Real_vi <0x1d2>;
defm V_MAX3_F32         : VOP3_Real_vi <0x1d3>;
defm V_MAX3_I32         : VOP3_Real_vi <0x1d4>;
defm V_MAX3_U32         : VOP3_Real_vi <0x1d5>;
defm V_MED3_F32         : VOP3_Real_vi <0x1d6>;
defm V_MED3_I32         : VOP3_Real_vi <0x1d7>;
defm V_MED3_U32         : VOP3_Real_vi <0x1d8>;
defm V_SAD_U8           : VOP3_Real_vi <0x1d9>;
defm V_SAD_HI_U8        : VOP3_Real_vi <0x1da>;
defm V_SAD_U16          : VOP3_Real_vi <0x1db>;
defm V_SAD_U32          : VOP3_Real_vi <0x1dc>;
defm V_CVT_PK_U8_F32    : VOP3_Real_vi <0x1dd>;
defm V_DIV_FIXUP_F32    : VOP3_Real_vi <0x1de>;
defm V_DIV_FIXUP_F64    : VOP3_Real_vi <0x1df>;
defm V_DIV_SCALE_F32    : VOP3be_Real_vi <0x1e0>;
defm V_DIV_SCALE_F64    : VOP3be_Real_vi <0x1e1>;
defm V_DIV_FMAS_F32     : VOP3_Real_vi <0x1e2>;
defm V_DIV_FMAS_F64     : VOP3_Real_vi <0x1e3>;
defm V_MSAD_U8          : VOP3_Real_vi <0x1e4>;
defm V_QSAD_PK_U16_U8   : VOP3_Real_vi <0x1e5>;
defm V_MQSAD_PK_U16_U8  : VOP3_Real_vi <0x1e6>;
defm V_MQSAD_U32_U8     : VOP3_Real_vi <0x1e7>;

defm V_MAD_F16          : VOP3_Real_vi <0x1ea>;
defm V_MAD_U16          : VOP3_Real_vi <0x1eb>;
defm V_MAD_I16          : VOP3_Real_vi <0x1ec>;

defm V_PERM_B32         : VOP3_Real_vi <0x1ed>;

defm V_FMA_F16          : VOP3_Real_vi <0x1ee>;
defm V_DIV_FIXUP_F16    : VOP3_Real_vi <0x1ef>;

defm V_INTERP_P1LL_F16  : VOP3_Real_vi <0x274>;
defm V_INTERP_P1LV_F16  : VOP3_Real_vi <0x275>;
defm V_INTERP_P2_F16    : VOP3_Real_vi <0x276>;
defm V_ADD_F64          : VOP3_Real_vi <0x280>;
defm V_MUL_F64          : VOP3_Real_vi <0x281>;
defm V_MIN_F64          : VOP3_Real_vi <0x282>;
defm V_MAX_F64          : VOP3_Real_vi <0x283>;
defm V_LDEXP_F64        : VOP3_Real_vi <0x284>;
defm V_MUL_LO_U32       : VOP3_Real_vi <0x285>;

// removed from VI as identical to V_MUL_LO_U32
let isAsmParserOnly = 1 in {
defm V_MUL_LO_I32       : VOP3_Real_vi <0x285>;
}

defm V_MUL_HI_U32       : VOP3_Real_vi <0x286>;
defm V_MUL_HI_I32       : VOP3_Real_vi <0x287>;

defm V_LSHLREV_B64      : VOP3_Real_vi <0x28f>;
defm V_LSHRREV_B64      : VOP3_Real_vi <0x290>;
defm V_ASHRREV_I64      : VOP3_Real_vi <0x291>;
defm V_TRIG_PREOP_F64   : VOP3_Real_vi <0x292>;

defm V_LSHL_ADD_U32 : VOP3_Real_vi <0x1fd>;
defm V_ADD_LSHL_U32 : VOP3_Real_vi <0x1fe>;
defm V_ADD3_U32 : VOP3_Real_vi <0x1ff>;
defm V_LSHL_OR_B32 : VOP3_Real_vi <0x200>;
defm V_AND_OR_B32 : VOP3_Real_vi <0x201>;
defm V_OR3_B32 : VOP3_Real_vi <0x202>;
defm V_PACK_B32_F16 : VOP3_Real_vi <0x2a0>;

defm V_XAD_U32 : VOP3_Real_vi <0x1f3>;

defm V_MIN3_F16 : VOP3_Real_vi <0x1f4>;
defm V_MIN3_I16 : VOP3_Real_vi <0x1f5>;
defm V_MIN3_U16 : VOP3_Real_vi <0x1f6>;

defm V_MAX3_F16 : VOP3_Real_vi <0x1f7>;
defm V_MAX3_I16 : VOP3_Real_vi <0x1f8>;
defm V_MAX3_U16 : VOP3_Real_vi <0x1f9>;

defm V_MED3_F16 : VOP3_Real_vi <0x1fa>;
defm V_MED3_I16 : VOP3_Real_vi <0x1fb>;
defm V_MED3_U16 : VOP3_Real_vi <0x1fc>;