1 //===- ConstantFold.cpp - LLVM constant folder ----------------------------===//
3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4 // See https://llvm.org/LICENSE.txt for license information.
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
7 //===----------------------------------------------------------------------===//
9 // This file implements folding of constants for LLVM. This implements the
10 // (internal) ConstantFold.h interface, which is used by the
11 // ConstantExpr::get* methods to automatically fold constants when possible.
13 // The current constant folding implementation is implemented in two pieces: the
14 // pieces that don't need DataLayout, and the pieces that do. This is to avoid
15 // a dependence in IR on Target.
17 //===----------------------------------------------------------------------===//
19 #include "llvm/IR/ConstantFold.h"
20 #include "llvm/ADT/APSInt.h"
21 #include "llvm/ADT/SmallVector.h"
22 #include "llvm/IR/Constants.h"
23 #include "llvm/IR/DerivedTypes.h"
24 #include "llvm/IR/Function.h"
25 #include "llvm/IR/GlobalAlias.h"
26 #include "llvm/IR/GlobalVariable.h"
27 #include "llvm/IR/Instructions.h"
28 #include "llvm/IR/Module.h"
29 #include "llvm/IR/Operator.h"
30 #include "llvm/IR/PatternMatch.h"
31 #include "llvm/Support/ErrorHandling.h"
33 using namespace llvm::PatternMatch
;
35 //===----------------------------------------------------------------------===//
36 // ConstantFold*Instruction Implementations
37 //===----------------------------------------------------------------------===//
39 /// This function determines which opcode to use to fold two constant cast
40 /// expressions together. It uses CastInst::isEliminableCastPair to determine
41 /// the opcode. Consequently its just a wrapper around that function.
42 /// Determine if it is valid to fold a cast of a cast
45 unsigned opc
, ///< opcode of the second cast constant expression
46 ConstantExpr
*Op
, ///< the first cast constant expression
47 Type
*DstTy
///< destination type of the first cast
49 assert(Op
&& Op
->isCast() && "Can't fold cast of cast without a cast!");
50 assert(DstTy
&& DstTy
->isFirstClassType() && "Invalid cast destination type");
51 assert(CastInst::isCast(opc
) && "Invalid cast opcode");
53 // The types and opcodes for the two Cast constant expressions
54 Type
*SrcTy
= Op
->getOperand(0)->getType();
55 Type
*MidTy
= Op
->getType();
56 Instruction::CastOps firstOp
= Instruction::CastOps(Op
->getOpcode());
57 Instruction::CastOps secondOp
= Instruction::CastOps(opc
);
59 // Assume that pointers are never more than 64 bits wide, and only use this
60 // for the middle type. Otherwise we could end up folding away illegal
61 // bitcasts between address spaces with different sizes.
62 IntegerType
*FakeIntPtrTy
= Type::getInt64Ty(DstTy
->getContext());
64 // Let CastInst::isEliminableCastPair do the heavy lifting.
65 return CastInst::isEliminableCastPair(firstOp
, secondOp
, SrcTy
, MidTy
, DstTy
,
66 nullptr, FakeIntPtrTy
, nullptr);
69 static Constant
*FoldBitCast(Constant
*V
, Type
*DestTy
) {
70 Type
*SrcTy
= V
->getType();
72 return V
; // no-op cast
74 if (V
->isAllOnesValue())
75 return Constant::getAllOnesValue(DestTy
);
77 // Handle ConstantInt -> ConstantFP
78 if (ConstantInt
*CI
= dyn_cast
<ConstantInt
>(V
)) {
79 // Canonicalize scalar-to-vector bitcasts into vector-to-vector bitcasts
80 // This allows for other simplifications (although some of them
81 // can only be handled by Analysis/ConstantFolding.cpp).
82 if (isa
<VectorType
>(DestTy
) && !isa
<VectorType
>(SrcTy
))
83 return ConstantExpr::getBitCast(ConstantVector::get(V
), DestTy
);
85 // Make sure dest type is compatible with the folded fp constant.
86 // See note below regarding the PPC_FP128 restriction.
87 if (!DestTy
->isFPOrFPVectorTy() || DestTy
->isPPC_FP128Ty() ||
88 DestTy
->getScalarSizeInBits() != SrcTy
->getScalarSizeInBits())
91 return ConstantFP::get(
93 APFloat(DestTy
->getScalarType()->getFltSemantics(), CI
->getValue()));
96 // Handle ConstantFP -> ConstantInt
97 if (ConstantFP
*FP
= dyn_cast
<ConstantFP
>(V
)) {
98 // Canonicalize scalar-to-vector bitcasts into vector-to-vector bitcasts
99 // This allows for other simplifications (although some of them
100 // can only be handled by Analysis/ConstantFolding.cpp).
101 if (isa
<VectorType
>(DestTy
) && !isa
<VectorType
>(SrcTy
))
102 return ConstantExpr::getBitCast(ConstantVector::get(V
), DestTy
);
104 // PPC_FP128 is really the sum of two consecutive doubles, where the first
105 // double is always stored first in memory, regardless of the target
106 // endianness. The memory layout of i128, however, depends on the target
107 // endianness, and so we can't fold this without target endianness
108 // information. This should instead be handled by
109 // Analysis/ConstantFolding.cpp
110 if (SrcTy
->isPPC_FP128Ty())
113 // Make sure dest type is compatible with the folded integer constant.
114 if (!DestTy
->isIntOrIntVectorTy() ||
115 DestTy
->getScalarSizeInBits() != SrcTy
->getScalarSizeInBits())
118 return ConstantInt::get(DestTy
, FP
->getValueAPF().bitcastToAPInt());
124 static Constant
*foldMaybeUndesirableCast(unsigned opc
, Constant
*V
,
126 return ConstantExpr::isDesirableCastOp(opc
)
127 ? ConstantExpr::getCast(opc
, V
, DestTy
)
128 : ConstantFoldCastInstruction(opc
, V
, DestTy
);
131 Constant
*llvm::ConstantFoldCastInstruction(unsigned opc
, Constant
*V
,
133 if (isa
<PoisonValue
>(V
))
134 return PoisonValue::get(DestTy
);
136 if (isa
<UndefValue
>(V
)) {
137 // zext(undef) = 0, because the top bits will be zero.
138 // sext(undef) = 0, because the top bits will all be the same.
139 // [us]itofp(undef) = 0, because the result value is bounded.
140 if (opc
== Instruction::ZExt
|| opc
== Instruction::SExt
||
141 opc
== Instruction::UIToFP
|| opc
== Instruction::SIToFP
)
142 return Constant::getNullValue(DestTy
);
143 return UndefValue::get(DestTy
);
146 if (V
->isNullValue() && !DestTy
->isX86_AMXTy() &&
147 opc
!= Instruction::AddrSpaceCast
)
148 return Constant::getNullValue(DestTy
);
150 // If the cast operand is a constant expression, there's a few things we can
151 // do to try to simplify it.
152 if (ConstantExpr
*CE
= dyn_cast
<ConstantExpr
>(V
)) {
154 // Try hard to fold cast of cast because they are often eliminable.
155 if (unsigned newOpc
= foldConstantCastPair(opc
, CE
, DestTy
))
156 return foldMaybeUndesirableCast(newOpc
, CE
->getOperand(0), DestTy
);
160 // If the cast operand is a constant vector, perform the cast by
161 // operating on each element. In the cast of bitcasts, the element
162 // count may be mismatched; don't attempt to handle that here.
163 if ((isa
<ConstantVector
>(V
) || isa
<ConstantDataVector
>(V
)) &&
164 DestTy
->isVectorTy() &&
165 cast
<FixedVectorType
>(DestTy
)->getNumElements() ==
166 cast
<FixedVectorType
>(V
->getType())->getNumElements()) {
167 VectorType
*DestVecTy
= cast
<VectorType
>(DestTy
);
168 Type
*DstEltTy
= DestVecTy
->getElementType();
169 // Fast path for splatted constants.
170 if (Constant
*Splat
= V
->getSplatValue()) {
171 Constant
*Res
= foldMaybeUndesirableCast(opc
, Splat
, DstEltTy
);
174 return ConstantVector::getSplat(
175 cast
<VectorType
>(DestTy
)->getElementCount(), Res
);
177 SmallVector
<Constant
*, 16> res
;
178 Type
*Ty
= IntegerType::get(V
->getContext(), 32);
180 e
= cast
<FixedVectorType
>(V
->getType())->getNumElements();
182 Constant
*C
= ConstantExpr::getExtractElement(V
, ConstantInt::get(Ty
, i
));
183 Constant
*Casted
= foldMaybeUndesirableCast(opc
, C
, DstEltTy
);
186 res
.push_back(Casted
);
188 return ConstantVector::get(res
);
191 // We actually have to do a cast now. Perform the cast according to the
195 llvm_unreachable("Failed to cast constant expression");
196 case Instruction::FPTrunc
:
197 case Instruction::FPExt
:
198 if (ConstantFP
*FPC
= dyn_cast
<ConstantFP
>(V
)) {
200 APFloat Val
= FPC
->getValueAPF();
201 Val
.convert(DestTy
->getScalarType()->getFltSemantics(),
202 APFloat::rmNearestTiesToEven
, &ignored
);
203 return ConstantFP::get(DestTy
, Val
);
205 return nullptr; // Can't fold.
206 case Instruction::FPToUI
:
207 case Instruction::FPToSI
:
208 if (ConstantFP
*FPC
= dyn_cast
<ConstantFP
>(V
)) {
209 const APFloat
&V
= FPC
->getValueAPF();
211 APSInt
IntVal(DestTy
->getScalarSizeInBits(), opc
== Instruction::FPToUI
);
212 if (APFloat::opInvalidOp
==
213 V
.convertToInteger(IntVal
, APFloat::rmTowardZero
, &ignored
)) {
214 // Undefined behavior invoked - the destination type can't represent
215 // the input constant.
216 return PoisonValue::get(DestTy
);
218 return ConstantInt::get(DestTy
, IntVal
);
220 return nullptr; // Can't fold.
221 case Instruction::UIToFP
:
222 case Instruction::SIToFP
:
223 if (ConstantInt
*CI
= dyn_cast
<ConstantInt
>(V
)) {
224 const APInt
&api
= CI
->getValue();
225 APFloat
apf(DestTy
->getScalarType()->getFltSemantics(),
226 APInt::getZero(DestTy
->getScalarSizeInBits()));
227 apf
.convertFromAPInt(api
, opc
==Instruction::SIToFP
,
228 APFloat::rmNearestTiesToEven
);
229 return ConstantFP::get(DestTy
, apf
);
232 case Instruction::ZExt
:
233 if (ConstantInt
*CI
= dyn_cast
<ConstantInt
>(V
)) {
234 uint32_t BitWidth
= DestTy
->getScalarSizeInBits();
235 return ConstantInt::get(DestTy
, CI
->getValue().zext(BitWidth
));
238 case Instruction::SExt
:
239 if (ConstantInt
*CI
= dyn_cast
<ConstantInt
>(V
)) {
240 uint32_t BitWidth
= DestTy
->getScalarSizeInBits();
241 return ConstantInt::get(DestTy
, CI
->getValue().sext(BitWidth
));
244 case Instruction::Trunc
: {
245 if (ConstantInt
*CI
= dyn_cast
<ConstantInt
>(V
)) {
246 uint32_t BitWidth
= DestTy
->getScalarSizeInBits();
247 return ConstantInt::get(DestTy
, CI
->getValue().trunc(BitWidth
));
252 case Instruction::BitCast
:
253 return FoldBitCast(V
, DestTy
);
254 case Instruction::AddrSpaceCast
:
255 case Instruction::IntToPtr
:
256 case Instruction::PtrToInt
:
261 Constant
*llvm::ConstantFoldSelectInstruction(Constant
*Cond
,
262 Constant
*V1
, Constant
*V2
) {
263 // Check for i1 and vector true/false conditions.
264 if (Cond
->isNullValue()) return V2
;
265 if (Cond
->isAllOnesValue()) return V1
;
267 // If the condition is a vector constant, fold the result elementwise.
268 if (ConstantVector
*CondV
= dyn_cast
<ConstantVector
>(Cond
)) {
269 auto *V1VTy
= CondV
->getType();
270 SmallVector
<Constant
*, 16> Result
;
271 Type
*Ty
= IntegerType::get(CondV
->getContext(), 32);
272 for (unsigned i
= 0, e
= V1VTy
->getNumElements(); i
!= e
; ++i
) {
274 Constant
*V1Element
= ConstantExpr::getExtractElement(V1
,
275 ConstantInt::get(Ty
, i
));
276 Constant
*V2Element
= ConstantExpr::getExtractElement(V2
,
277 ConstantInt::get(Ty
, i
));
278 auto *Cond
= cast
<Constant
>(CondV
->getOperand(i
));
279 if (isa
<PoisonValue
>(Cond
)) {
280 V
= PoisonValue::get(V1Element
->getType());
281 } else if (V1Element
== V2Element
) {
283 } else if (isa
<UndefValue
>(Cond
)) {
284 V
= isa
<UndefValue
>(V1Element
) ? V1Element
: V2Element
;
286 if (!isa
<ConstantInt
>(Cond
)) break;
287 V
= Cond
->isNullValue() ? V2Element
: V1Element
;
292 // If we were able to build the vector, return it.
293 if (Result
.size() == V1VTy
->getNumElements())
294 return ConstantVector::get(Result
);
297 if (isa
<PoisonValue
>(Cond
))
298 return PoisonValue::get(V1
->getType());
300 if (isa
<UndefValue
>(Cond
)) {
301 if (isa
<UndefValue
>(V1
)) return V1
;
305 if (V1
== V2
) return V1
;
307 if (isa
<PoisonValue
>(V1
))
309 if (isa
<PoisonValue
>(V2
))
312 // If the true or false value is undef, we can fold to the other value as
313 // long as the other value isn't poison.
314 auto NotPoison
= [](Constant
*C
) {
315 if (isa
<PoisonValue
>(C
))
318 // TODO: We can analyze ConstExpr by opcode to determine if there is any
319 // possibility of poison.
320 if (isa
<ConstantExpr
>(C
))
323 if (isa
<ConstantInt
>(C
) || isa
<GlobalVariable
>(C
) || isa
<ConstantFP
>(C
) ||
324 isa
<ConstantPointerNull
>(C
) || isa
<Function
>(C
))
327 if (C
->getType()->isVectorTy())
328 return !C
->containsPoisonElement() && !C
->containsConstantExpression();
330 // TODO: Recursively analyze aggregates or other constants.
333 if (isa
<UndefValue
>(V1
) && NotPoison(V2
)) return V2
;
334 if (isa
<UndefValue
>(V2
) && NotPoison(V1
)) return V1
;
339 Constant
*llvm::ConstantFoldExtractElementInstruction(Constant
*Val
,
341 auto *ValVTy
= cast
<VectorType
>(Val
->getType());
343 // extractelt poison, C -> poison
344 // extractelt C, undef -> poison
345 if (isa
<PoisonValue
>(Val
) || isa
<UndefValue
>(Idx
))
346 return PoisonValue::get(ValVTy
->getElementType());
348 // extractelt undef, C -> undef
349 if (isa
<UndefValue
>(Val
))
350 return UndefValue::get(ValVTy
->getElementType());
352 auto *CIdx
= dyn_cast
<ConstantInt
>(Idx
);
356 if (auto *ValFVTy
= dyn_cast
<FixedVectorType
>(Val
->getType())) {
357 // ee({w,x,y,z}, wrong_value) -> poison
358 if (CIdx
->uge(ValFVTy
->getNumElements()))
359 return PoisonValue::get(ValFVTy
->getElementType());
362 // ee (gep (ptr, idx0, ...), idx) -> gep (ee (ptr, idx), ee (idx0, idx), ...)
363 if (auto *CE
= dyn_cast
<ConstantExpr
>(Val
)) {
364 if (auto *GEP
= dyn_cast
<GEPOperator
>(CE
)) {
365 SmallVector
<Constant
*, 8> Ops
;
366 Ops
.reserve(CE
->getNumOperands());
367 for (unsigned i
= 0, e
= CE
->getNumOperands(); i
!= e
; ++i
) {
368 Constant
*Op
= CE
->getOperand(i
);
369 if (Op
->getType()->isVectorTy()) {
370 Constant
*ScalarOp
= ConstantExpr::getExtractElement(Op
, Idx
);
373 Ops
.push_back(ScalarOp
);
377 return CE
->getWithOperands(Ops
, ValVTy
->getElementType(), false,
378 GEP
->getSourceElementType());
379 } else if (CE
->getOpcode() == Instruction::InsertElement
) {
380 if (const auto *IEIdx
= dyn_cast
<ConstantInt
>(CE
->getOperand(2))) {
381 if (APSInt::isSameValue(APSInt(IEIdx
->getValue()),
382 APSInt(CIdx
->getValue()))) {
383 return CE
->getOperand(1);
385 return ConstantExpr::getExtractElement(CE
->getOperand(0), CIdx
);
391 if (Constant
*C
= Val
->getAggregateElement(CIdx
))
394 // Lane < Splat minimum vector width => extractelt Splat(x), Lane -> x
395 if (CIdx
->getValue().ult(ValVTy
->getElementCount().getKnownMinValue())) {
396 if (Constant
*SplatVal
= Val
->getSplatValue())
403 Constant
*llvm::ConstantFoldInsertElementInstruction(Constant
*Val
,
406 if (isa
<UndefValue
>(Idx
))
407 return PoisonValue::get(Val
->getType());
409 // Inserting null into all zeros is still all zeros.
410 // TODO: This is true for undef and poison splats too.
411 if (isa
<ConstantAggregateZero
>(Val
) && Elt
->isNullValue())
414 ConstantInt
*CIdx
= dyn_cast
<ConstantInt
>(Idx
);
415 if (!CIdx
) return nullptr;
417 // Do not iterate on scalable vector. The num of elements is unknown at
419 if (isa
<ScalableVectorType
>(Val
->getType()))
422 auto *ValTy
= cast
<FixedVectorType
>(Val
->getType());
424 unsigned NumElts
= ValTy
->getNumElements();
425 if (CIdx
->uge(NumElts
))
426 return PoisonValue::get(Val
->getType());
428 SmallVector
<Constant
*, 16> Result
;
429 Result
.reserve(NumElts
);
430 auto *Ty
= Type::getInt32Ty(Val
->getContext());
431 uint64_t IdxVal
= CIdx
->getZExtValue();
432 for (unsigned i
= 0; i
!= NumElts
; ++i
) {
434 Result
.push_back(Elt
);
438 Constant
*C
= ConstantExpr::getExtractElement(Val
, ConstantInt::get(Ty
, i
));
442 return ConstantVector::get(Result
);
445 Constant
*llvm::ConstantFoldShuffleVectorInstruction(Constant
*V1
, Constant
*V2
,
446 ArrayRef
<int> Mask
) {
447 auto *V1VTy
= cast
<VectorType
>(V1
->getType());
448 unsigned MaskNumElts
= Mask
.size();
450 ElementCount::get(MaskNumElts
, isa
<ScalableVectorType
>(V1VTy
));
451 Type
*EltTy
= V1VTy
->getElementType();
453 // Poison shuffle mask -> poison value.
454 if (all_of(Mask
, [](int Elt
) { return Elt
== PoisonMaskElem
; })) {
455 return PoisonValue::get(VectorType::get(EltTy
, MaskEltCount
));
458 // If the mask is all zeros this is a splat, no need to go through all
460 if (all_of(Mask
, [](int Elt
) { return Elt
== 0; })) {
461 Type
*Ty
= IntegerType::get(V1
->getContext(), 32);
463 ConstantExpr::getExtractElement(V1
, ConstantInt::get(Ty
, 0));
465 if (Elt
->isNullValue()) {
466 auto *VTy
= VectorType::get(EltTy
, MaskEltCount
);
467 return ConstantAggregateZero::get(VTy
);
468 } else if (!MaskEltCount
.isScalable())
469 return ConstantVector::getSplat(MaskEltCount
, Elt
);
472 // Do not iterate on scalable vector. The num of elements is unknown at
474 if (isa
<ScalableVectorType
>(V1VTy
))
477 unsigned SrcNumElts
= V1VTy
->getElementCount().getKnownMinValue();
479 // Loop over the shuffle mask, evaluating each element.
480 SmallVector
<Constant
*, 32> Result
;
481 for (unsigned i
= 0; i
!= MaskNumElts
; ++i
) {
484 Result
.push_back(UndefValue::get(EltTy
));
488 if (unsigned(Elt
) >= SrcNumElts
*2)
489 InElt
= UndefValue::get(EltTy
);
490 else if (unsigned(Elt
) >= SrcNumElts
) {
491 Type
*Ty
= IntegerType::get(V2
->getContext(), 32);
493 ConstantExpr::getExtractElement(V2
,
494 ConstantInt::get(Ty
, Elt
- SrcNumElts
));
496 Type
*Ty
= IntegerType::get(V1
->getContext(), 32);
497 InElt
= ConstantExpr::getExtractElement(V1
, ConstantInt::get(Ty
, Elt
));
499 Result
.push_back(InElt
);
502 return ConstantVector::get(Result
);
505 Constant
*llvm::ConstantFoldExtractValueInstruction(Constant
*Agg
,
506 ArrayRef
<unsigned> Idxs
) {
507 // Base case: no indices, so return the entire value.
511 if (Constant
*C
= Agg
->getAggregateElement(Idxs
[0]))
512 return ConstantFoldExtractValueInstruction(C
, Idxs
.slice(1));
517 Constant
*llvm::ConstantFoldInsertValueInstruction(Constant
*Agg
,
519 ArrayRef
<unsigned> Idxs
) {
520 // Base case: no indices, so replace the entire value.
525 if (StructType
*ST
= dyn_cast
<StructType
>(Agg
->getType()))
526 NumElts
= ST
->getNumElements();
528 NumElts
= cast
<ArrayType
>(Agg
->getType())->getNumElements();
530 SmallVector
<Constant
*, 32> Result
;
531 for (unsigned i
= 0; i
!= NumElts
; ++i
) {
532 Constant
*C
= Agg
->getAggregateElement(i
);
533 if (!C
) return nullptr;
536 C
= ConstantFoldInsertValueInstruction(C
, Val
, Idxs
.slice(1));
541 if (StructType
*ST
= dyn_cast
<StructType
>(Agg
->getType()))
542 return ConstantStruct::get(ST
, Result
);
543 return ConstantArray::get(cast
<ArrayType
>(Agg
->getType()), Result
);
546 Constant
*llvm::ConstantFoldUnaryInstruction(unsigned Opcode
, Constant
*C
) {
547 assert(Instruction::isUnaryOp(Opcode
) && "Non-unary instruction detected");
549 // Handle scalar UndefValue and scalable vector UndefValue. Fixed-length
550 // vectors are always evaluated per element.
551 bool IsScalableVector
= isa
<ScalableVectorType
>(C
->getType());
552 bool HasScalarUndefOrScalableVectorUndef
=
553 (!C
->getType()->isVectorTy() || IsScalableVector
) && isa
<UndefValue
>(C
);
555 if (HasScalarUndefOrScalableVectorUndef
) {
556 switch (static_cast<Instruction::UnaryOps
>(Opcode
)) {
557 case Instruction::FNeg
:
558 return C
; // -undef -> undef
559 case Instruction::UnaryOpsEnd
:
560 llvm_unreachable("Invalid UnaryOp");
564 // Constant should not be UndefValue, unless these are vector constants.
565 assert(!HasScalarUndefOrScalableVectorUndef
&& "Unexpected UndefValue");
566 // We only have FP UnaryOps right now.
567 assert(!isa
<ConstantInt
>(C
) && "Unexpected Integer UnaryOp");
569 if (ConstantFP
*CFP
= dyn_cast
<ConstantFP
>(C
)) {
570 const APFloat
&CV
= CFP
->getValueAPF();
574 case Instruction::FNeg
:
575 return ConstantFP::get(C
->getType(), neg(CV
));
577 } else if (auto *VTy
= dyn_cast
<VectorType
>(C
->getType())) {
578 // Fast path for splatted constants.
579 if (Constant
*Splat
= C
->getSplatValue())
580 if (Constant
*Elt
= ConstantFoldUnaryInstruction(Opcode
, Splat
))
581 return ConstantVector::getSplat(VTy
->getElementCount(), Elt
);
583 if (auto *FVTy
= dyn_cast
<FixedVectorType
>(VTy
)) {
584 // Fold each element and create a vector constant from those constants.
585 Type
*Ty
= IntegerType::get(FVTy
->getContext(), 32);
586 SmallVector
<Constant
*, 16> Result
;
587 for (unsigned i
= 0, e
= FVTy
->getNumElements(); i
!= e
; ++i
) {
588 Constant
*ExtractIdx
= ConstantInt::get(Ty
, i
);
589 Constant
*Elt
= ConstantExpr::getExtractElement(C
, ExtractIdx
);
590 Constant
*Res
= ConstantFoldUnaryInstruction(Opcode
, Elt
);
593 Result
.push_back(Res
);
596 return ConstantVector::get(Result
);
600 // We don't know how to fold this.
604 Constant
*llvm::ConstantFoldBinaryInstruction(unsigned Opcode
, Constant
*C1
,
606 assert(Instruction::isBinaryOp(Opcode
) && "Non-binary instruction detected");
608 // Simplify BinOps with their identity values first. They are no-ops and we
609 // can always return the other value, including undef or poison values.
610 if (Constant
*Identity
= ConstantExpr::getBinOpIdentity(
611 Opcode
, C1
->getType(), /*AllowRHSIdentity*/ false)) {
616 } else if (Constant
*Identity
= ConstantExpr::getBinOpIdentity(
617 Opcode
, C1
->getType(), /*AllowRHSIdentity*/ true)) {
622 // Binary operations propagate poison.
623 if (isa
<PoisonValue
>(C1
) || isa
<PoisonValue
>(C2
))
624 return PoisonValue::get(C1
->getType());
626 // Handle scalar UndefValue and scalable vector UndefValue. Fixed-length
627 // vectors are always evaluated per element.
628 bool IsScalableVector
= isa
<ScalableVectorType
>(C1
->getType());
629 bool HasScalarUndefOrScalableVectorUndef
=
630 (!C1
->getType()->isVectorTy() || IsScalableVector
) &&
631 (isa
<UndefValue
>(C1
) || isa
<UndefValue
>(C2
));
632 if (HasScalarUndefOrScalableVectorUndef
) {
633 switch (static_cast<Instruction::BinaryOps
>(Opcode
)) {
634 case Instruction::Xor
:
635 if (isa
<UndefValue
>(C1
) && isa
<UndefValue
>(C2
))
636 // Handle undef ^ undef -> 0 special case. This is a common
638 return Constant::getNullValue(C1
->getType());
640 case Instruction::Add
:
641 case Instruction::Sub
:
642 return UndefValue::get(C1
->getType());
643 case Instruction::And
:
644 if (isa
<UndefValue
>(C1
) && isa
<UndefValue
>(C2
)) // undef & undef -> undef
646 return Constant::getNullValue(C1
->getType()); // undef & X -> 0
647 case Instruction::Mul
: {
648 // undef * undef -> undef
649 if (isa
<UndefValue
>(C1
) && isa
<UndefValue
>(C2
))
652 // X * undef -> undef if X is odd
653 if (match(C1
, m_APInt(CV
)) || match(C2
, m_APInt(CV
)))
655 return UndefValue::get(C1
->getType());
657 // X * undef -> 0 otherwise
658 return Constant::getNullValue(C1
->getType());
660 case Instruction::SDiv
:
661 case Instruction::UDiv
:
662 // X / undef -> poison
664 if (match(C2
, m_CombineOr(m_Undef(), m_Zero())))
665 return PoisonValue::get(C2
->getType());
666 // undef / X -> 0 otherwise
667 return Constant::getNullValue(C1
->getType());
668 case Instruction::URem
:
669 case Instruction::SRem
:
670 // X % undef -> poison
672 if (match(C2
, m_CombineOr(m_Undef(), m_Zero())))
673 return PoisonValue::get(C2
->getType());
674 // undef % X -> 0 otherwise
675 return Constant::getNullValue(C1
->getType());
676 case Instruction::Or
: // X | undef -> -1
677 if (isa
<UndefValue
>(C1
) && isa
<UndefValue
>(C2
)) // undef | undef -> undef
679 return Constant::getAllOnesValue(C1
->getType()); // undef | X -> ~0
680 case Instruction::LShr
:
681 // X >>l undef -> poison
682 if (isa
<UndefValue
>(C2
))
683 return PoisonValue::get(C2
->getType());
685 return Constant::getNullValue(C1
->getType());
686 case Instruction::AShr
:
687 // X >>a undef -> poison
688 if (isa
<UndefValue
>(C2
))
689 return PoisonValue::get(C2
->getType());
690 // TODO: undef >>a X -> poison if the shift is exact
692 return Constant::getNullValue(C1
->getType());
693 case Instruction::Shl
:
694 // X << undef -> undef
695 if (isa
<UndefValue
>(C2
))
696 return PoisonValue::get(C2
->getType());
698 return Constant::getNullValue(C1
->getType());
699 case Instruction::FSub
:
700 // -0.0 - undef --> undef (consistent with "fneg undef")
701 if (match(C1
, m_NegZeroFP()) && isa
<UndefValue
>(C2
))
704 case Instruction::FAdd
:
705 case Instruction::FMul
:
706 case Instruction::FDiv
:
707 case Instruction::FRem
:
708 // [any flop] undef, undef -> undef
709 if (isa
<UndefValue
>(C1
) && isa
<UndefValue
>(C2
))
711 // [any flop] C, undef -> NaN
712 // [any flop] undef, C -> NaN
713 // We could potentially specialize NaN/Inf constants vs. 'normal'
714 // constants (possibly differently depending on opcode and operand). This
715 // would allow returning undef sometimes. But it is always safe to fold to
716 // NaN because we can choose the undef operand as NaN, and any FP opcode
717 // with a NaN operand will propagate NaN.
718 return ConstantFP::getNaN(C1
->getType());
719 case Instruction::BinaryOpsEnd
:
720 llvm_unreachable("Invalid BinaryOp");
724 // Neither constant should be UndefValue, unless these are vector constants.
725 assert((!HasScalarUndefOrScalableVectorUndef
) && "Unexpected UndefValue");
727 // Handle simplifications when the RHS is a constant int.
728 if (ConstantInt
*CI2
= dyn_cast
<ConstantInt
>(C2
)) {
729 if (C2
== ConstantExpr::getBinOpAbsorber(Opcode
, C2
->getType(),
730 /*AllowLHSConstant*/ false))
734 case Instruction::UDiv
:
735 case Instruction::SDiv
:
737 return PoisonValue::get(CI2
->getType()); // X / 0 == poison
739 case Instruction::URem
:
740 case Instruction::SRem
:
742 return Constant::getNullValue(CI2
->getType()); // X % 1 == 0
744 return PoisonValue::get(CI2
->getType()); // X % 0 == poison
746 case Instruction::And
:
747 assert(!CI2
->isZero() && "And zero handled above");
748 if (ConstantExpr
*CE1
= dyn_cast
<ConstantExpr
>(C1
)) {
749 // If and'ing the address of a global with a constant, fold it.
750 if (CE1
->getOpcode() == Instruction::PtrToInt
&&
751 isa
<GlobalValue
>(CE1
->getOperand(0))) {
752 GlobalValue
*GV
= cast
<GlobalValue
>(CE1
->getOperand(0));
754 Align GVAlign
; // defaults to 1
756 if (Module
*TheModule
= GV
->getParent()) {
757 const DataLayout
&DL
= TheModule
->getDataLayout();
758 GVAlign
= GV
->getPointerAlignment(DL
);
760 // If the function alignment is not specified then assume that it
762 // This is dangerous; on x86, the alignment of the pointer
763 // corresponds to the alignment of the function, but might be less
764 // than 4 if it isn't explicitly specified.
765 // However, a fix for this behaviour was reverted because it
766 // increased code size (see https://reviews.llvm.org/D55115)
767 // FIXME: This code should be deleted once existing targets have
768 // appropriate defaults
769 if (isa
<Function
>(GV
) && !DL
.getFunctionPtrAlign())
771 } else if (isa
<GlobalVariable
>(GV
)) {
772 GVAlign
= cast
<GlobalVariable
>(GV
)->getAlign().valueOrOne();
776 unsigned DstWidth
= CI2
->getBitWidth();
777 unsigned SrcWidth
= std::min(DstWidth
, Log2(GVAlign
));
778 APInt
BitsNotSet(APInt::getLowBitsSet(DstWidth
, SrcWidth
));
780 // If checking bits we know are clear, return zero.
781 if ((CI2
->getValue() & BitsNotSet
) == CI2
->getValue())
782 return Constant::getNullValue(CI2
->getType());
788 } else if (isa
<ConstantInt
>(C1
)) {
789 // If C1 is a ConstantInt and C2 is not, swap the operands.
790 if (Instruction::isCommutative(Opcode
))
791 return ConstantExpr::isDesirableBinOp(Opcode
)
792 ? ConstantExpr::get(Opcode
, C2
, C1
)
793 : ConstantFoldBinaryInstruction(Opcode
, C2
, C1
);
796 if (ConstantInt
*CI1
= dyn_cast
<ConstantInt
>(C1
)) {
797 if (ConstantInt
*CI2
= dyn_cast
<ConstantInt
>(C2
)) {
798 const APInt
&C1V
= CI1
->getValue();
799 const APInt
&C2V
= CI2
->getValue();
803 case Instruction::Add
:
804 return ConstantInt::get(C1
->getType(), C1V
+ C2V
);
805 case Instruction::Sub
:
806 return ConstantInt::get(C1
->getType(), C1V
- C2V
);
807 case Instruction::Mul
:
808 return ConstantInt::get(C1
->getType(), C1V
* C2V
);
809 case Instruction::UDiv
:
810 assert(!CI2
->isZero() && "Div by zero handled above");
811 return ConstantInt::get(CI1
->getType(), C1V
.udiv(C2V
));
812 case Instruction::SDiv
:
813 assert(!CI2
->isZero() && "Div by zero handled above");
814 if (C2V
.isAllOnes() && C1V
.isMinSignedValue())
815 return PoisonValue::get(CI1
->getType()); // MIN_INT / -1 -> poison
816 return ConstantInt::get(CI1
->getType(), C1V
.sdiv(C2V
));
817 case Instruction::URem
:
818 assert(!CI2
->isZero() && "Div by zero handled above");
819 return ConstantInt::get(C1
->getType(), C1V
.urem(C2V
));
820 case Instruction::SRem
:
821 assert(!CI2
->isZero() && "Div by zero handled above");
822 if (C2V
.isAllOnes() && C1V
.isMinSignedValue())
823 return PoisonValue::get(C1
->getType()); // MIN_INT % -1 -> poison
824 return ConstantInt::get(C1
->getType(), C1V
.srem(C2V
));
825 case Instruction::And
:
826 return ConstantInt::get(C1
->getType(), C1V
& C2V
);
827 case Instruction::Or
:
828 return ConstantInt::get(C1
->getType(), C1V
| C2V
);
829 case Instruction::Xor
:
830 return ConstantInt::get(C1
->getType(), C1V
^ C2V
);
831 case Instruction::Shl
:
832 if (C2V
.ult(C1V
.getBitWidth()))
833 return ConstantInt::get(C1
->getType(), C1V
.shl(C2V
));
834 return PoisonValue::get(C1
->getType()); // too big shift is poison
835 case Instruction::LShr
:
836 if (C2V
.ult(C1V
.getBitWidth()))
837 return ConstantInt::get(C1
->getType(), C1V
.lshr(C2V
));
838 return PoisonValue::get(C1
->getType()); // too big shift is poison
839 case Instruction::AShr
:
840 if (C2V
.ult(C1V
.getBitWidth()))
841 return ConstantInt::get(C1
->getType(), C1V
.ashr(C2V
));
842 return PoisonValue::get(C1
->getType()); // too big shift is poison
846 if (C1
== ConstantExpr::getBinOpAbsorber(Opcode
, C1
->getType(),
847 /*AllowLHSConstant*/ true))
849 } else if (ConstantFP
*CFP1
= dyn_cast
<ConstantFP
>(C1
)) {
850 if (ConstantFP
*CFP2
= dyn_cast
<ConstantFP
>(C2
)) {
851 const APFloat
&C1V
= CFP1
->getValueAPF();
852 const APFloat
&C2V
= CFP2
->getValueAPF();
853 APFloat C3V
= C1V
; // copy for modification
857 case Instruction::FAdd
:
858 (void)C3V
.add(C2V
, APFloat::rmNearestTiesToEven
);
859 return ConstantFP::get(C1
->getType(), C3V
);
860 case Instruction::FSub
:
861 (void)C3V
.subtract(C2V
, APFloat::rmNearestTiesToEven
);
862 return ConstantFP::get(C1
->getType(), C3V
);
863 case Instruction::FMul
:
864 (void)C3V
.multiply(C2V
, APFloat::rmNearestTiesToEven
);
865 return ConstantFP::get(C1
->getType(), C3V
);
866 case Instruction::FDiv
:
867 (void)C3V
.divide(C2V
, APFloat::rmNearestTiesToEven
);
868 return ConstantFP::get(C1
->getType(), C3V
);
869 case Instruction::FRem
:
871 return ConstantFP::get(C1
->getType(), C3V
);
876 if (auto *VTy
= dyn_cast
<VectorType
>(C1
->getType())) {
877 // Fast path for splatted constants.
878 if (Constant
*C2Splat
= C2
->getSplatValue()) {
879 if (Instruction::isIntDivRem(Opcode
) && C2Splat
->isNullValue())
880 return PoisonValue::get(VTy
);
881 if (Constant
*C1Splat
= C1
->getSplatValue()) {
883 ConstantExpr::isDesirableBinOp(Opcode
)
884 ? ConstantExpr::get(Opcode
, C1Splat
, C2Splat
)
885 : ConstantFoldBinaryInstruction(Opcode
, C1Splat
, C2Splat
);
888 return ConstantVector::getSplat(VTy
->getElementCount(), Res
);
892 if (auto *FVTy
= dyn_cast
<FixedVectorType
>(VTy
)) {
893 // Fold each element and create a vector constant from those constants.
894 SmallVector
<Constant
*, 16> Result
;
895 Type
*Ty
= IntegerType::get(FVTy
->getContext(), 32);
896 for (unsigned i
= 0, e
= FVTy
->getNumElements(); i
!= e
; ++i
) {
897 Constant
*ExtractIdx
= ConstantInt::get(Ty
, i
);
898 Constant
*LHS
= ConstantExpr::getExtractElement(C1
, ExtractIdx
);
899 Constant
*RHS
= ConstantExpr::getExtractElement(C2
, ExtractIdx
);
900 Constant
*Res
= ConstantExpr::isDesirableBinOp(Opcode
)
901 ? ConstantExpr::get(Opcode
, LHS
, RHS
)
902 : ConstantFoldBinaryInstruction(Opcode
, LHS
, RHS
);
905 Result
.push_back(Res
);
908 return ConstantVector::get(Result
);
912 if (ConstantExpr
*CE1
= dyn_cast
<ConstantExpr
>(C1
)) {
913 // There are many possible foldings we could do here. We should probably
914 // at least fold add of a pointer with an integer into the appropriate
915 // getelementptr. This will improve alias analysis a bit.
917 // Given ((a + b) + c), if (b + c) folds to something interesting, return
919 if (Instruction::isAssociative(Opcode
) && CE1
->getOpcode() == Opcode
) {
920 Constant
*T
= ConstantExpr::get(Opcode
, CE1
->getOperand(1), C2
);
921 if (!isa
<ConstantExpr
>(T
) || cast
<ConstantExpr
>(T
)->getOpcode() != Opcode
)
922 return ConstantExpr::get(Opcode
, CE1
->getOperand(0), T
);
924 } else if (isa
<ConstantExpr
>(C2
)) {
925 // If C2 is a constant expr and C1 isn't, flop them around and fold the
926 // other way if possible.
927 if (Instruction::isCommutative(Opcode
))
928 return ConstantFoldBinaryInstruction(Opcode
, C2
, C1
);
931 // i1 can be simplified in many cases.
932 if (C1
->getType()->isIntegerTy(1)) {
934 case Instruction::Add
:
935 case Instruction::Sub
:
936 return ConstantExpr::getXor(C1
, C2
);
937 case Instruction::Shl
:
938 case Instruction::LShr
:
939 case Instruction::AShr
:
940 // We can assume that C2 == 0. If it were one the result would be
941 // undefined because the shift value is as large as the bitwidth.
943 case Instruction::SDiv
:
944 case Instruction::UDiv
:
945 // We can assume that C2 == 1. If it were zero the result would be
946 // undefined through division by zero.
948 case Instruction::URem
:
949 case Instruction::SRem
:
950 // We can assume that C2 == 1. If it were zero the result would be
951 // undefined through division by zero.
952 return ConstantInt::getFalse(C1
->getContext());
958 // We don't know how to fold this.
962 static ICmpInst::Predicate
areGlobalsPotentiallyEqual(const GlobalValue
*GV1
,
963 const GlobalValue
*GV2
) {
964 auto isGlobalUnsafeForEquality
= [](const GlobalValue
*GV
) {
965 if (GV
->isInterposable() || GV
->hasGlobalUnnamedAddr())
967 if (const auto *GVar
= dyn_cast
<GlobalVariable
>(GV
)) {
968 Type
*Ty
= GVar
->getValueType();
969 // A global with opaque type might end up being zero sized.
972 // A global with an empty type might lie at the address of any other
979 // Don't try to decide equality of aliases.
980 if (!isa
<GlobalAlias
>(GV1
) && !isa
<GlobalAlias
>(GV2
))
981 if (!isGlobalUnsafeForEquality(GV1
) && !isGlobalUnsafeForEquality(GV2
))
982 return ICmpInst::ICMP_NE
;
983 return ICmpInst::BAD_ICMP_PREDICATE
;
986 /// This function determines if there is anything we can decide about the two
987 /// constants provided. This doesn't need to handle simple things like integer
988 /// comparisons, but should instead handle ConstantExprs and GlobalValues.
989 /// If we can determine that the two constants have a particular relation to
990 /// each other, we should return the corresponding ICmp predicate, otherwise
991 /// return ICmpInst::BAD_ICMP_PREDICATE.
992 static ICmpInst::Predicate
evaluateICmpRelation(Constant
*V1
, Constant
*V2
) {
993 assert(V1
->getType() == V2
->getType() &&
994 "Cannot compare different types of values!");
995 if (V1
== V2
) return ICmpInst::ICMP_EQ
;
997 // The following folds only apply to pointers.
998 if (!V1
->getType()->isPointerTy())
999 return ICmpInst::BAD_ICMP_PREDICATE
;
1001 // To simplify this code we canonicalize the relation so that the first
1002 // operand is always the most "complex" of the two. We consider simple
1003 // constants (like ConstantPointerNull) to be the simplest, followed by
1004 // BlockAddress, GlobalValues, and ConstantExpr's (the most complex).
1005 auto GetComplexity
= [](Constant
*V
) {
1006 if (isa
<ConstantExpr
>(V
))
1008 if (isa
<GlobalValue
>(V
))
1010 if (isa
<BlockAddress
>(V
))
1014 if (GetComplexity(V1
) < GetComplexity(V2
)) {
1015 ICmpInst::Predicate SwappedRelation
= evaluateICmpRelation(V2
, V1
);
1016 if (SwappedRelation
!= ICmpInst::BAD_ICMP_PREDICATE
)
1017 return ICmpInst::getSwappedPredicate(SwappedRelation
);
1018 return ICmpInst::BAD_ICMP_PREDICATE
;
1021 if (const BlockAddress
*BA
= dyn_cast
<BlockAddress
>(V1
)) {
1022 // Now we know that the RHS is a BlockAddress or simple constant.
1023 if (const BlockAddress
*BA2
= dyn_cast
<BlockAddress
>(V2
)) {
1024 // Block address in another function can't equal this one, but block
1025 // addresses in the current function might be the same if blocks are
1027 if (BA2
->getFunction() != BA
->getFunction())
1028 return ICmpInst::ICMP_NE
;
1029 } else if (isa
<ConstantPointerNull
>(V2
)) {
1030 return ICmpInst::ICMP_NE
;
1032 } else if (const GlobalValue
*GV
= dyn_cast
<GlobalValue
>(V1
)) {
1033 // Now we know that the RHS is a GlobalValue, BlockAddress or simple
1035 if (const GlobalValue
*GV2
= dyn_cast
<GlobalValue
>(V2
)) {
1036 return areGlobalsPotentiallyEqual(GV
, GV2
);
1037 } else if (isa
<BlockAddress
>(V2
)) {
1038 return ICmpInst::ICMP_NE
; // Globals never equal labels.
1039 } else if (isa
<ConstantPointerNull
>(V2
)) {
1040 // GlobalVals can never be null unless they have external weak linkage.
1041 // We don't try to evaluate aliases here.
1042 // NOTE: We should not be doing this constant folding if null pointer
1043 // is considered valid for the function. But currently there is no way to
1044 // query it from the Constant type.
1045 if (!GV
->hasExternalWeakLinkage() && !isa
<GlobalAlias
>(GV
) &&
1046 !NullPointerIsDefined(nullptr /* F */,
1047 GV
->getType()->getAddressSpace()))
1048 return ICmpInst::ICMP_UGT
;
1050 } else if (auto *CE1
= dyn_cast
<ConstantExpr
>(V1
)) {
1051 // Ok, the LHS is known to be a constantexpr. The RHS can be any of a
1052 // constantexpr, a global, block address, or a simple constant.
1053 Constant
*CE1Op0
= CE1
->getOperand(0);
1055 switch (CE1
->getOpcode()) {
1056 case Instruction::GetElementPtr
: {
1057 GEPOperator
*CE1GEP
= cast
<GEPOperator
>(CE1
);
1058 // Ok, since this is a getelementptr, we know that the constant has a
1059 // pointer type. Check the various cases.
1060 if (isa
<ConstantPointerNull
>(V2
)) {
1061 // If we are comparing a GEP to a null pointer, check to see if the base
1062 // of the GEP equals the null pointer.
1063 if (const GlobalValue
*GV
= dyn_cast
<GlobalValue
>(CE1Op0
)) {
1064 // If its not weak linkage, the GVal must have a non-zero address
1065 // so the result is greater-than
1066 if (!GV
->hasExternalWeakLinkage() && CE1GEP
->isInBounds())
1067 return ICmpInst::ICMP_UGT
;
1069 } else if (const GlobalValue
*GV2
= dyn_cast
<GlobalValue
>(V2
)) {
1070 if (const GlobalValue
*GV
= dyn_cast
<GlobalValue
>(CE1Op0
)) {
1072 if (CE1GEP
->hasAllZeroIndices())
1073 return areGlobalsPotentiallyEqual(GV
, GV2
);
1074 return ICmpInst::BAD_ICMP_PREDICATE
;
1077 } else if (const auto *CE2GEP
= dyn_cast
<GEPOperator
>(V2
)) {
1078 // By far the most common case to handle is when the base pointers are
1079 // obviously to the same global.
1080 const Constant
*CE2Op0
= cast
<Constant
>(CE2GEP
->getPointerOperand());
1081 if (isa
<GlobalValue
>(CE1Op0
) && isa
<GlobalValue
>(CE2Op0
)) {
1082 // Don't know relative ordering, but check for inequality.
1083 if (CE1Op0
!= CE2Op0
) {
1084 if (CE1GEP
->hasAllZeroIndices() && CE2GEP
->hasAllZeroIndices())
1085 return areGlobalsPotentiallyEqual(cast
<GlobalValue
>(CE1Op0
),
1086 cast
<GlobalValue
>(CE2Op0
));
1087 return ICmpInst::BAD_ICMP_PREDICATE
;
1098 return ICmpInst::BAD_ICMP_PREDICATE
;
1101 Constant
*llvm::ConstantFoldCompareInstruction(CmpInst::Predicate Predicate
,
1102 Constant
*C1
, Constant
*C2
) {
1104 if (VectorType
*VT
= dyn_cast
<VectorType
>(C1
->getType()))
1105 ResultTy
= VectorType::get(Type::getInt1Ty(C1
->getContext()),
1106 VT
->getElementCount());
1108 ResultTy
= Type::getInt1Ty(C1
->getContext());
1110 // Fold FCMP_FALSE/FCMP_TRUE unconditionally.
1111 if (Predicate
== FCmpInst::FCMP_FALSE
)
1112 return Constant::getNullValue(ResultTy
);
1114 if (Predicate
== FCmpInst::FCMP_TRUE
)
1115 return Constant::getAllOnesValue(ResultTy
);
1117 // Handle some degenerate cases first
1118 if (isa
<PoisonValue
>(C1
) || isa
<PoisonValue
>(C2
))
1119 return PoisonValue::get(ResultTy
);
1121 if (isa
<UndefValue
>(C1
) || isa
<UndefValue
>(C2
)) {
1122 bool isIntegerPredicate
= ICmpInst::isIntPredicate(Predicate
);
1123 // For EQ and NE, we can always pick a value for the undef to make the
1124 // predicate pass or fail, so we can return undef.
1125 // Also, if both operands are undef, we can return undef for int comparison.
1126 if (ICmpInst::isEquality(Predicate
) || (isIntegerPredicate
&& C1
== C2
))
1127 return UndefValue::get(ResultTy
);
1129 // Otherwise, for integer compare, pick the same value as the non-undef
1130 // operand, and fold it to true or false.
1131 if (isIntegerPredicate
)
1132 return ConstantInt::get(ResultTy
, CmpInst::isTrueWhenEqual(Predicate
));
1134 // Choosing NaN for the undef will always make unordered comparison succeed
1135 // and ordered comparison fails.
1136 return ConstantInt::get(ResultTy
, CmpInst::isUnordered(Predicate
));
1139 if (C2
->isNullValue()) {
1140 // The caller is expected to commute the operands if the constant expression
1143 if (Predicate
== ICmpInst::ICMP_UGE
)
1144 return Constant::getAllOnesValue(ResultTy
);
1146 if (Predicate
== ICmpInst::ICMP_ULT
)
1147 return Constant::getNullValue(ResultTy
);
1150 // If the comparison is a comparison between two i1's, simplify it.
1151 if (C1
->getType()->isIntegerTy(1)) {
1152 switch (Predicate
) {
1153 case ICmpInst::ICMP_EQ
:
1154 if (isa
<ConstantInt
>(C2
))
1155 return ConstantExpr::getXor(C1
, ConstantExpr::getNot(C2
));
1156 return ConstantExpr::getXor(ConstantExpr::getNot(C1
), C2
);
1157 case ICmpInst::ICMP_NE
:
1158 return ConstantExpr::getXor(C1
, C2
);
1164 if (isa
<ConstantInt
>(C1
) && isa
<ConstantInt
>(C2
)) {
1165 const APInt
&V1
= cast
<ConstantInt
>(C1
)->getValue();
1166 const APInt
&V2
= cast
<ConstantInt
>(C2
)->getValue();
1167 return ConstantInt::get(ResultTy
, ICmpInst::compare(V1
, V2
, Predicate
));
1168 } else if (isa
<ConstantFP
>(C1
) && isa
<ConstantFP
>(C2
)) {
1169 const APFloat
&C1V
= cast
<ConstantFP
>(C1
)->getValueAPF();
1170 const APFloat
&C2V
= cast
<ConstantFP
>(C2
)->getValueAPF();
1171 return ConstantInt::get(ResultTy
, FCmpInst::compare(C1V
, C2V
, Predicate
));
1172 } else if (auto *C1VTy
= dyn_cast
<VectorType
>(C1
->getType())) {
1174 // Fast path for splatted constants.
1175 if (Constant
*C1Splat
= C1
->getSplatValue())
1176 if (Constant
*C2Splat
= C2
->getSplatValue())
1178 ConstantFoldCompareInstruction(Predicate
, C1Splat
, C2Splat
))
1179 return ConstantVector::getSplat(C1VTy
->getElementCount(), Elt
);
1181 // Do not iterate on scalable vector. The number of elements is unknown at
1183 if (isa
<ScalableVectorType
>(C1VTy
))
1186 // If we can constant fold the comparison of each element, constant fold
1187 // the whole vector comparison.
1188 SmallVector
<Constant
*, 4> ResElts
;
1189 Type
*Ty
= IntegerType::get(C1
->getContext(), 32);
1190 // Compare the elements, producing an i1 result or constant expr.
1191 for (unsigned I
= 0, E
= C1VTy
->getElementCount().getKnownMinValue();
1194 ConstantExpr::getExtractElement(C1
, ConstantInt::get(Ty
, I
));
1196 ConstantExpr::getExtractElement(C2
, ConstantInt::get(Ty
, I
));
1197 Constant
*Elt
= ConstantFoldCompareInstruction(Predicate
, C1E
, C2E
);
1201 ResElts
.push_back(Elt
);
1204 return ConstantVector::get(ResElts
);
1207 if (C1
->getType()->isFPOrFPVectorTy()) {
1209 // We know that C1 == C2 || isUnordered(C1, C2).
1210 if (Predicate
== FCmpInst::FCMP_ONE
)
1211 return ConstantInt::getFalse(ResultTy
);
1212 else if (Predicate
== FCmpInst::FCMP_UEQ
)
1213 return ConstantInt::getTrue(ResultTy
);
1216 // Evaluate the relation between the two constants, per the predicate.
1217 int Result
= -1; // -1 = unknown, 0 = known false, 1 = known true.
1218 switch (evaluateICmpRelation(C1
, C2
)) {
1219 default: llvm_unreachable("Unknown relational!");
1220 case ICmpInst::BAD_ICMP_PREDICATE
:
1221 break; // Couldn't determine anything about these constants.
1222 case ICmpInst::ICMP_EQ
: // We know the constants are equal!
1223 // If we know the constants are equal, we can decide the result of this
1224 // computation precisely.
1225 Result
= ICmpInst::isTrueWhenEqual(Predicate
);
1227 case ICmpInst::ICMP_ULT
:
1228 switch (Predicate
) {
1229 case ICmpInst::ICMP_ULT
: case ICmpInst::ICMP_NE
: case ICmpInst::ICMP_ULE
:
1231 case ICmpInst::ICMP_UGT
: case ICmpInst::ICMP_EQ
: case ICmpInst::ICMP_UGE
:
1237 case ICmpInst::ICMP_SLT
:
1238 switch (Predicate
) {
1239 case ICmpInst::ICMP_SLT
: case ICmpInst::ICMP_NE
: case ICmpInst::ICMP_SLE
:
1241 case ICmpInst::ICMP_SGT
: case ICmpInst::ICMP_EQ
: case ICmpInst::ICMP_SGE
:
1247 case ICmpInst::ICMP_UGT
:
1248 switch (Predicate
) {
1249 case ICmpInst::ICMP_UGT
: case ICmpInst::ICMP_NE
: case ICmpInst::ICMP_UGE
:
1251 case ICmpInst::ICMP_ULT
: case ICmpInst::ICMP_EQ
: case ICmpInst::ICMP_ULE
:
1257 case ICmpInst::ICMP_SGT
:
1258 switch (Predicate
) {
1259 case ICmpInst::ICMP_SGT
: case ICmpInst::ICMP_NE
: case ICmpInst::ICMP_SGE
:
1261 case ICmpInst::ICMP_SLT
: case ICmpInst::ICMP_EQ
: case ICmpInst::ICMP_SLE
:
1267 case ICmpInst::ICMP_ULE
:
1268 if (Predicate
== ICmpInst::ICMP_UGT
)
1270 if (Predicate
== ICmpInst::ICMP_ULT
|| Predicate
== ICmpInst::ICMP_ULE
)
1273 case ICmpInst::ICMP_SLE
:
1274 if (Predicate
== ICmpInst::ICMP_SGT
)
1276 if (Predicate
== ICmpInst::ICMP_SLT
|| Predicate
== ICmpInst::ICMP_SLE
)
1279 case ICmpInst::ICMP_UGE
:
1280 if (Predicate
== ICmpInst::ICMP_ULT
)
1282 if (Predicate
== ICmpInst::ICMP_UGT
|| Predicate
== ICmpInst::ICMP_UGE
)
1285 case ICmpInst::ICMP_SGE
:
1286 if (Predicate
== ICmpInst::ICMP_SLT
)
1288 if (Predicate
== ICmpInst::ICMP_SGT
|| Predicate
== ICmpInst::ICMP_SGE
)
1291 case ICmpInst::ICMP_NE
:
1292 if (Predicate
== ICmpInst::ICMP_EQ
)
1294 if (Predicate
== ICmpInst::ICMP_NE
)
1299 // If we evaluated the result, return it now.
1301 return ConstantInt::get(ResultTy
, Result
);
1303 if ((!isa
<ConstantExpr
>(C1
) && isa
<ConstantExpr
>(C2
)) ||
1304 (C1
->isNullValue() && !C2
->isNullValue())) {
1305 // If C2 is a constant expr and C1 isn't, flip them around and fold the
1306 // other way if possible.
1307 // Also, if C1 is null and C2 isn't, flip them around.
1308 Predicate
= ICmpInst::getSwappedPredicate(Predicate
);
1309 return ConstantFoldCompareInstruction(Predicate
, C2
, C1
);
1315 Constant
*llvm::ConstantFoldGetElementPtr(Type
*PointeeTy
, Constant
*C
,
1316 std::optional
<ConstantRange
> InRange
,
1317 ArrayRef
<Value
*> Idxs
) {
1318 if (Idxs
.empty()) return C
;
1320 Type
*GEPTy
= GetElementPtrInst::getGEPReturnType(
1321 C
, ArrayRef((Value
*const *)Idxs
.data(), Idxs
.size()));
1323 if (isa
<PoisonValue
>(C
))
1324 return PoisonValue::get(GEPTy
);
1326 if (isa
<UndefValue
>(C
))
1327 return UndefValue::get(GEPTy
);
1329 auto IsNoOp
= [&]() {
1330 // Avoid losing inrange information.
1334 return all_of(Idxs
, [](Value
*Idx
) {
1335 Constant
*IdxC
= cast
<Constant
>(Idx
);
1336 return IdxC
->isNullValue() || isa
<UndefValue
>(IdxC
);
1340 return GEPTy
->isVectorTy() && !C
->getType()->isVectorTy()
1341 ? ConstantVector::getSplat(
1342 cast
<VectorType
>(GEPTy
)->getElementCount(), C
)