summaryrefslogtreecommitdiffstats
path: root/src/gallium/drivers/radeon/R600ISelLowering.cpp
blob: ec1250d14da634d9a079488eb989731b041c770d (plain)
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
//===-- R600ISelLowering.cpp - R600 DAG Lowering Implementation -----------===//
//
//                     The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// Most of the DAG lowering is handled in AMDGPUISelLowering.cpp.  This file
// is mostly EmitInstrWithCustomInserter().
//
//===----------------------------------------------------------------------===//

#include "R600ISelLowering.h"
#include "R600InstrInfo.h"
#include "R600MachineFunctionInfo.h"
#include "llvm/CodeGen/MachineInstrBuilder.h"
#include "llvm/CodeGen/MachineRegisterInfo.h"
#include "llvm/CodeGen/SelectionDAG.h"

using namespace llvm;

R600TargetLowering::R600TargetLowering(TargetMachine &TM) :
    AMDGPUTargetLowering(TM),
    TII(static_cast<const R600InstrInfo*>(TM.getInstrInfo()))
{
  setOperationAction(ISD::MUL, MVT::i64, Expand);
  addRegisterClass(MVT::v4f32, &AMDGPU::R600_Reg128RegClass);
  addRegisterClass(MVT::f32, &AMDGPU::R600_Reg32RegClass);
  addRegisterClass(MVT::v4i32, &AMDGPU::R600_Reg128RegClass);
  addRegisterClass(MVT::i32, &AMDGPU::R600_Reg32RegClass);
  computeRegisterProperties();

  setOperationAction(ISD::BR_CC, MVT::i32, Custom);

  setOperationAction(ISD::FSUB, MVT::f32, Expand);

  setOperationAction(ISD::INTRINSIC_VOID, MVT::Other, Custom);
  setOperationAction(ISD::INTRINSIC_WO_CHAIN, MVT::Other, Custom);

  setOperationAction(ISD::ROTL, MVT::i32, Custom);

  setOperationAction(ISD::SELECT_CC, MVT::f32, Custom);
  setOperationAction(ISD::SELECT_CC, MVT::i32, Custom);

  setOperationAction(ISD::SETCC, MVT::i32, Custom);

  setSchedulingPreference(Sched::VLIW);
}

MachineBasicBlock * R600TargetLowering::EmitInstrWithCustomInserter(
    MachineInstr * MI, MachineBasicBlock * BB) const
{
  MachineFunction * MF = BB->getParent();
  MachineRegisterInfo &MRI = MF->getRegInfo();
  MachineBasicBlock::iterator I = *MI;

  switch (MI->getOpcode()) {
  default: return AMDGPUTargetLowering::EmitInstrWithCustomInserter(MI, BB);
  case AMDGPU::CLAMP_R600:
    {
      MachineInstr *NewMI =
        BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::MOV))
               .addOperand(MI->getOperand(0))
               .addOperand(MI->getOperand(1))
               .addReg(AMDGPU::PRED_SEL_OFF);
      TII->AddFlag(NewMI, 0, MO_FLAG_CLAMP);
      break;
    }
  case AMDGPU::FABS_R600:
    {
      MachineInstr *NewMI =
        BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::MOV))
               .addOperand(MI->getOperand(0))
               .addOperand(MI->getOperand(1))
               .addReg(AMDGPU::PRED_SEL_OFF);
      TII->AddFlag(NewMI, 1, MO_FLAG_ABS);
      break;
    }

  case AMDGPU::FNEG_R600:
    {
      MachineInstr *NewMI =
        BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::MOV))
                .addOperand(MI->getOperand(0))
                .addOperand(MI->getOperand(1))
                .addReg(AMDGPU::PRED_SEL_OFF);
      TII->AddFlag(NewMI, 1, MO_FLAG_NEG);
    break;
    }

  case AMDGPU::R600_LOAD_CONST:
    {
      int64_t RegIndex = MI->getOperand(1).getImm();
      unsigned ConstantReg = AMDGPU::R600_CReg32RegClass.getRegister(RegIndex);
      BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::COPY))
                  .addOperand(MI->getOperand(0))
                  .addReg(ConstantReg);
      break;
    }

  case AMDGPU::MASK_WRITE:
    {
      unsigned maskedRegister = MI->getOperand(0).getReg();
      assert(TargetRegisterInfo::isVirtualRegister(maskedRegister));
      MachineInstr * defInstr = MRI.getVRegDef(maskedRegister);
      TII->AddFlag(defInstr, 0, MO_FLAG_MASK);
      // Return early so the instruction is not erased
      return BB;
    }

  case AMDGPU::RAT_WRITE_CACHELESS_eg:
    {
      // Convert to DWORD address
      unsigned NewAddr = MRI.createVirtualRegister(
                                             AMDGPU::R600_TReg32_XRegisterClass);
      unsigned ShiftValue = MRI.createVirtualRegister(
                                              AMDGPU::R600_TReg32RegisterClass);

      // XXX In theory, we should be able to pass ShiftValue directly to
      // the LSHR_eg instruction as an inline literal, but I tried doing it
      // this way and it didn't produce the correct results.
      BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::MOV), ShiftValue)
              .addReg(AMDGPU::ALU_LITERAL_X)
              .addReg(AMDGPU::PRED_SEL_OFF)
              .addImm(2);
      BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::LSHR_eg), NewAddr)
              .addOperand(MI->getOperand(1))
              .addReg(ShiftValue)
              .addReg(AMDGPU::PRED_SEL_OFF);
      BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(MI->getOpcode()))
              .addOperand(MI->getOperand(0))
              .addReg(NewAddr);
      break;
    }

  case AMDGPU::RESERVE_REG:
    {
      R600MachineFunctionInfo * MFI = MF->getInfo<R600MachineFunctionInfo>();
      int64_t ReservedIndex = MI->getOperand(0).getImm();
      unsigned ReservedReg =
                          AMDGPU::R600_TReg32RegClass.getRegister(ReservedIndex);
      MFI->ReservedRegs.push_back(ReservedReg);
      break;
    }

  case AMDGPU::TXD:
    {
      unsigned t0 = MRI.createVirtualRegister(AMDGPU::R600_Reg128RegisterClass);
      unsigned t1 = MRI.createVirtualRegister(AMDGPU::R600_Reg128RegisterClass);

      BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::TEX_SET_GRADIENTS_H), t0)
              .addOperand(MI->getOperand(3))
              .addOperand(MI->getOperand(4))
              .addOperand(MI->getOperand(5));
      BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::TEX_SET_GRADIENTS_V), t1)
              .addOperand(MI->getOperand(2))
              .addOperand(MI->getOperand(4))
              .addOperand(MI->getOperand(5));
      BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::TEX_SAMPLE_G))
              .addOperand(MI->getOperand(0))
              .addOperand(MI->getOperand(1))
              .addOperand(MI->getOperand(4))
              .addOperand(MI->getOperand(5))
              .addReg(t0, RegState::Implicit)
              .addReg(t1, RegState::Implicit);
      break;
    }
  case AMDGPU::TXD_SHADOW:
    {
      unsigned t0 = MRI.createVirtualRegister(AMDGPU::R600_Reg128RegisterClass);
      unsigned t1 = MRI.createVirtualRegister(AMDGPU::R600_Reg128RegisterClass);

      BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::TEX_SET_GRADIENTS_H), t0)
              .addOperand(MI->getOperand(3))
              .addOperand(MI->getOperand(4))
              .addOperand(MI->getOperand(5));
      BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::TEX_SET_GRADIENTS_V), t1)
              .addOperand(MI->getOperand(2))
              .addOperand(MI->getOperand(4))
              .addOperand(MI->getOperand(5));
      BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::TEX_SAMPLE_C_G))
              .addOperand(MI->getOperand(0))
              .addOperand(MI->getOperand(1))
              .addOperand(MI->getOperand(4))
              .addOperand(MI->getOperand(5))
              .addReg(t0, RegState::Implicit)
              .addReg(t1, RegState::Implicit);
      break;
    }
  case AMDGPU::BRANCH:
      BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::JUMP))
              .addOperand(MI->getOperand(0))
              .addReg(0);
      break;
  case AMDGPU::BRANCH_COND_f32:
    {
      MachineInstr *NewMI =
        BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::PRED_X))
                .addReg(AMDGPU::PREDICATE_BIT)
                .addOperand(MI->getOperand(1))
                .addImm(OPCODE_IS_ZERO);
      TII->AddFlag(NewMI, 1, MO_FLAG_PUSH);
      BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::JUMP))
              .addOperand(MI->getOperand(0))
              .addReg(AMDGPU::PREDICATE_BIT, RegState::Kill);
      break;
    }
  case AMDGPU::BRANCH_COND_i32:
    {
      MachineInstr *NewMI =
        BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::PRED_X))
              .addReg(AMDGPU::PREDICATE_BIT)
              .addOperand(MI->getOperand(1))
              .addImm(OPCODE_IS_ZERO_INT);
      TII->AddFlag(NewMI, 1, MO_FLAG_PUSH);
      BuildMI(*BB, I, BB->findDebugLoc(I), TII->get(AMDGPU::JUMP))
             .addOperand(MI->getOperand(0))
              .addReg(AMDGPU::PREDICATE_BIT, RegState::Kill);
      break;
    }
  }

  MI->eraseFromParent();
  return BB;
}

//===----------------------------------------------------------------------===//
// Custom DAG Lowering Operations
//===----------------------------------------------------------------------===//

using namespace llvm::Intrinsic;
using namespace llvm::AMDGPUIntrinsic;

SDValue R600TargetLowering::LowerOperation(SDValue Op, SelectionDAG &DAG) const
{
  switch (Op.getOpcode()) {
  default: return AMDGPUTargetLowering::LowerOperation(Op, DAG);
  case ISD::BR_CC: return LowerBR_CC(Op, DAG);
  case ISD::ROTL: return LowerROTL(Op, DAG);
  case ISD::SELECT_CC: return LowerSELECT_CC(Op, DAG);
  case ISD::SETCC: return LowerSETCC(Op, DAG);
  case ISD::INTRINSIC_VOID: {
    SDValue Chain = Op.getOperand(0);
    unsigned IntrinsicID =
                         cast<ConstantSDNode>(Op.getOperand(1))->getZExtValue();
    switch (IntrinsicID) {
    case AMDGPUIntrinsic::AMDGPU_store_output: {
      MachineFunction &MF = DAG.getMachineFunction();
      MachineRegisterInfo &MRI = MF.getRegInfo();
      int64_t RegIndex = cast<ConstantSDNode>(Op.getOperand(3))->getZExtValue();
      unsigned Reg = AMDGPU::R600_TReg32RegClass.getRegister(RegIndex);
      if (!MRI.isLiveOut(Reg)) {
        MRI.addLiveOut(Reg);
      }
      return DAG.getCopyToReg(Chain, Op.getDebugLoc(), Reg, Op.getOperand(2));
    }
    // default for switch(IntrinsicID)
    default: break;
    }
    // break out of case ISD::INTRINSIC_VOID in switch(Op.getOpcode())
    break;
  }
  case ISD::INTRINSIC_WO_CHAIN: {
    unsigned IntrinsicID =
                         cast<ConstantSDNode>(Op.getOperand(0))->getZExtValue();
    EVT VT = Op.getValueType();
    DebugLoc DL = Op.getDebugLoc();
    switch(IntrinsicID) {
    default: return AMDGPUTargetLowering::LowerOperation(Op, DAG);
    case AMDGPUIntrinsic::R600_load_input: {
      int64_t RegIndex = cast<ConstantSDNode>(Op.getOperand(1))->getZExtValue();
      unsigned Reg = AMDGPU::R600_TReg32RegClass.getRegister(RegIndex);
      return CreateLiveInRegister(DAG, &AMDGPU::R600_TReg32RegClass, Reg, VT);
    }

    case r600_read_ngroups_x:
      return LowerImplicitParameter(DAG, VT, DL, 0);
    case r600_read_ngroups_y:
      return LowerImplicitParameter(DAG, VT, DL, 1);
    case r600_read_ngroups_z:
      return LowerImplicitParameter(DAG, VT, DL, 2);
    case r600_read_global_size_x:
      return LowerImplicitParameter(DAG, VT, DL, 3);
    case r600_read_global_size_y:
      return LowerImplicitParameter(DAG, VT, DL, 4);
    case r600_read_global_size_z:
      return LowerImplicitParameter(DAG, VT, DL, 5);
    case r600_read_local_size_x:
      return LowerImplicitParameter(DAG, VT, DL, 6);
    case r600_read_local_size_y:
      return LowerImplicitParameter(DAG, VT, DL, 7);
    case r600_read_local_size_z:
      return LowerImplicitParameter(DAG, VT, DL, 8);

    case r600_read_tgid_x:
      return CreateLiveInRegister(DAG, &AMDGPU::R600_TReg32RegClass,
                                  AMDGPU::T1_X, VT);
    case r600_read_tgid_y:
      return CreateLiveInRegister(DAG, &AMDGPU::R600_TReg32RegClass,
                                  AMDGPU::T1_Y, VT);
    case r600_read_tgid_z:
      return CreateLiveInRegister(DAG, &AMDGPU::R600_TReg32RegClass,
                                  AMDGPU::T1_Z, VT);
    case r600_read_tidig_x:
      return CreateLiveInRegister(DAG, &AMDGPU::R600_TReg32RegClass,
                                  AMDGPU::T0_X, VT);
    case r600_read_tidig_y:
      return CreateLiveInRegister(DAG, &AMDGPU::R600_TReg32RegClass,
                                  AMDGPU::T0_Y, VT);
    case r600_read_tidig_z:
      return CreateLiveInRegister(DAG, &AMDGPU::R600_TReg32RegClass,
                                  AMDGPU::T0_Z, VT);
    }
    // break out of case ISD::INTRINSIC_WO_CHAIN in switch(Op.getOpcode())
    break;
  }
  } // end switch(Op.getOpcode())
  return SDValue();
}

SDValue R600TargetLowering::LowerBR_CC(SDValue Op, SelectionDAG &DAG) const
{
  SDValue Chain = Op.getOperand(0);
  SDValue CC = Op.getOperand(1);
  SDValue LHS   = Op.getOperand(2);
  SDValue RHS   = Op.getOperand(3);
  SDValue JumpT  = Op.getOperand(4);
  SDValue CmpValue;
  SDValue Result;
  CmpValue = DAG.getNode(
      ISD::SELECT_CC,
      Op.getDebugLoc(),
      MVT::i32,
      LHS, RHS,
      DAG.getConstant(-1, MVT::i32),
      DAG.getConstant(0, MVT::i32),
      CC);
  Result = DAG.getNode(
      AMDGPUISD::BRANCH_COND,
      CmpValue.getDebugLoc(),
      MVT::Other, Chain,
      JumpT, CmpValue);
  return Result;
}

SDValue R600TargetLowering::LowerImplicitParameter(SelectionDAG &DAG, EVT VT,
                                                   DebugLoc DL,
                                                   unsigned DwordOffset) const
{
  unsigned ByteOffset = DwordOffset * 4;
  PointerType * PtrType = PointerType::get(VT.getTypeForEVT(*DAG.getContext()),
                                      AMDGPUAS::PARAM_I_ADDRESS);

  // We shouldn't be using an offset wider than 16-bits for implicit parameters.
  assert(isInt<16>(ByteOffset));

  return DAG.getLoad(VT, DL, DAG.getEntryNode(),
                     DAG.getConstant(ByteOffset, MVT::i32), // PTR
                     MachinePointerInfo(ConstantPointerNull::get(PtrType)),
                     false, false, false, 0);
}

SDValue R600TargetLowering::LowerROTL(SDValue Op, SelectionDAG &DAG) const
{
  DebugLoc DL = Op.getDebugLoc();
  EVT VT = Op.getValueType();

  return DAG.getNode(AMDGPUISD::BITALIGN, DL, VT,
                     Op.getOperand(0),
                     Op.getOperand(0),
                     DAG.getNode(ISD::SUB, DL, VT,
                                 DAG.getConstant(32, MVT::i32),
                                 Op.getOperand(1)));
}

SDValue R600TargetLowering::LowerSELECT_CC(SDValue Op, SelectionDAG &DAG) const
{
  DebugLoc DL = Op.getDebugLoc();
  EVT VT = Op.getValueType();

  SDValue LHS = Op.getOperand(0);
  SDValue RHS = Op.getOperand(1);
  SDValue True = Op.getOperand(2);
  SDValue False = Op.getOperand(3);
  SDValue CC = Op.getOperand(4);
  ISD::CondCode CCOpcode = cast<CondCodeSDNode>(CC)->get();
  SDValue Temp;

  // LHS and RHS are guaranteed to be the same value type
  EVT CompareVT = LHS.getValueType();

  // We need all the operands of SELECT_CC to have the same value type, so if
  // necessary we need to convert LHS and RHS to be the same type True and
  // False.  True and False are guaranteed to have the same type as this
  // SELECT_CC node.

  if (CompareVT !=  VT) {
    ISD::NodeType ConversionOp = ISD::DELETED_NODE;
    if (VT == MVT::f32 && CompareVT == MVT::i32) {
      if (isUnsignedIntSetCC(CCOpcode)) {
        ConversionOp = ISD::UINT_TO_FP;
      } else {
        ConversionOp = ISD::SINT_TO_FP;
      }
    } else if (VT == MVT::i32 && CompareVT == MVT::f32) {
      ConversionOp = ISD::FP_TO_SINT;
    } else {
      // I don't think there will be any other type pairings.
      assert(!"Unhandled operand type parings in SELECT_CC");
    }
    // XXX Check the value of LHS and RHS and avoid creating sequences like
    // (FTOI (ITOF))
    LHS = DAG.getNode(ConversionOp, DL, VT, LHS);
    RHS = DAG.getNode(ConversionOp, DL, VT, RHS);
  }

  // If True is a hardware TRUE value and False is a hardware FALSE value or
  // vice-versa we can handle this with a native instruction (SET* instructions).
  if ((isHWTrueValue(True) && isHWFalseValue(False))) {
    return DAG.getNode(ISD::SELECT_CC, DL, VT, LHS, RHS, True, False, CC);
  }

  // XXX If True is a hardware TRUE value and False is a hardware FALSE value,
  // we can handle this with a native instruction, but we need to swap true
  // and false and change the conditional.
  if (isHWTrueValue(False) && isHWFalseValue(True)) {
  }

  // XXX Check if we can lower this to a SELECT or if it is supported by a native
  // operation. (The code below does this but we don't have the Instruction
  // selection patterns to do this yet.
#if 0
  if (isZero(LHS) || isZero(RHS)) {
    SDValue Cond = (isZero(LHS) ? RHS : LHS);
    bool SwapTF = false;
    switch (CCOpcode) {
    case ISD::SETOEQ:
    case ISD::SETUEQ:
    case ISD::SETEQ:
      SwapTF = true;
      // Fall through
    case ISD::SETONE:
    case ISD::SETUNE:
    case ISD::SETNE:
      // We can lower to select
      if (SwapTF) {
        Temp = True;
        True = False;
        False = Temp;
      }
      // CNDE
      return DAG.getNode(ISD::SELECT, DL, VT, Cond, True, False);
    default:
      // Supported by a native operation (CNDGE, CNDGT)
      return DAG.getNode(ISD::SELECT_CC, DL, VT, LHS, RHS, True, False, CC);
    }
  }
#endif

  // If we make it this for it means we have no native instructions to handle
  // this SELECT_CC, so we must lower it.
  SDValue HWTrue, HWFalse;

  if (VT == MVT::f32) {
    HWTrue = DAG.getConstantFP(1.0f, VT);
    HWFalse = DAG.getConstantFP(0.0f, VT);
  } else if (VT == MVT::i32) {
    HWTrue = DAG.getConstant(-1, VT);
    HWFalse = DAG.getConstant(0, VT);
  }
  else {
    assert(!"Unhandled value type in LowerSELECT_CC");
  }

  // Lower this unsupported SELECT_CC into a combination of two supported
  // SELECT_CC operations.
  SDValue Cond = DAG.getNode(ISD::SELECT_CC, DL, VT, LHS, RHS, HWTrue, HWFalse, CC);

  // Convert floating point condition to i1
  if (VT == MVT::f32) {
    Cond = DAG.getNode(ISD::FP_TO_SINT, DL, MVT::i32,
                       DAG.getNode(ISD::FNEG, DL, VT, Cond));
  }

  return DAG.getNode(ISD::SELECT, DL, VT, Cond, True, False);
}

SDValue R600TargetLowering::LowerSETCC(SDValue Op, SelectionDAG &DAG) const
{
  SDValue Cond;
  SDValue LHS = Op.getOperand(0);
  SDValue RHS = Op.getOperand(1);
  SDValue CC  = Op.getOperand(2);
  DebugLoc DL = Op.getDebugLoc();
  assert(Op.getValueType() == MVT::i32);
  Cond = DAG.getNode(
      ISD::SELECT_CC,
      Op.getDebugLoc(),
      MVT::i32,
      LHS, RHS,
      DAG.getConstant(-1, MVT::i32),
      DAG.getConstant(0, MVT::i32),
      CC);
  Cond = DAG.getNode(
      ISD::AND,
      DL,
      MVT::i32,
      DAG.getConstant(1, MVT::i32),
      Cond);
  return Cond;
}