aaron/ispc
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
Files
326c45fa171654930760995cc2bc2f719eb116b6
ispc/llvmutil.cpp
Matt Pharr 326c45fa17 Fix bugs in LLVMExtractFirstVectorElement(). 
When we're manually scalarizing the extraction of the first element
of a vector value, we need to be careful about handling constant values
and about where new instructions are inserted.  The old code was
sloppy about this, which in turn lead to invalid IR in some cases.
For example, the two bugs below were essentially due to generating
an extractelement inst from a zeroinitializer value and then inserting
it in the wrong bblock such that a phi node that used that value was
malformed.

Fixes issues #240 and #229.
2012-04-19 09:45:04 -07:00

1555 lines
54 KiB
C++
Raw Blame History

/*
Copyright (c) 2010-2012, Intel Corporation
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.
* Neither the name of Intel Corporation nor the names of its
contributors may be used to endorse or promote products derived from
this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS
IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER
OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
/** @file llvmutil.cpp
@brief Implementations of various LLVM utility types and classes.
*/
#include "llvmutil.h"
#include "ispc.h"
#include "type.h"
#include <llvm/Instructions.h>
#include <llvm/BasicBlock.h>
#include <set>
#include <map>
llvm::Type *LLVMTypes::VoidType = NULL;
llvm::PointerType *LLVMTypes::VoidPointerType = NULL;
llvm::Type *LLVMTypes::PointerIntType = NULL;
llvm::Type *LLVMTypes::BoolType = NULL;
llvm::Type *LLVMTypes::Int8Type = NULL;
llvm::Type *LLVMTypes::Int16Type = NULL;
llvm::Type *LLVMTypes::Int32Type = NULL;
llvm::Type *LLVMTypes::Int64Type = NULL;
llvm::Type *LLVMTypes::FloatType = NULL;
llvm::Type *LLVMTypes::DoubleType = NULL;
llvm::Type *LLVMTypes::Int8PointerType = NULL;
llvm::Type *LLVMTypes::Int16PointerType = NULL;
llvm::Type *LLVMTypes::Int32PointerType = NULL;
llvm::Type *LLVMTypes::Int64PointerType = NULL;
llvm::Type *LLVMTypes::FloatPointerType = NULL;
llvm::Type *LLVMTypes::DoublePointerType = NULL;
llvm::VectorType *LLVMTypes::MaskType = NULL;
llvm::VectorType *LLVMTypes::BoolVectorType = NULL;
llvm::VectorType *LLVMTypes::Int1VectorType = NULL;
llvm::VectorType *LLVMTypes::Int8VectorType = NULL;
llvm::VectorType *LLVMTypes::Int16VectorType = NULL;
llvm::VectorType *LLVMTypes::Int32VectorType = NULL;
llvm::VectorType *LLVMTypes::Int64VectorType = NULL;
llvm::VectorType *LLVMTypes::FloatVectorType = NULL;
llvm::VectorType *LLVMTypes::DoubleVectorType = NULL;
llvm::Type *LLVMTypes::Int8VectorPointerType = NULL;
llvm::Type *LLVMTypes::Int16VectorPointerType = NULL;
llvm::Type *LLVMTypes::Int32VectorPointerType = NULL;
llvm::Type *LLVMTypes::Int64VectorPointerType = NULL;
llvm::Type *LLVMTypes::FloatVectorPointerType = NULL;
llvm::Type *LLVMTypes::DoubleVectorPointerType = NULL;
llvm::VectorType *LLVMTypes::VoidPointerVectorType = NULL;
llvm::Constant *LLVMTrue = NULL;
llvm::Constant *LLVMFalse = NULL;
llvm::Constant *LLVMMaskAllOn = NULL;
llvm::Constant *LLVMMaskAllOff = NULL;
void
InitLLVMUtil(llvm::LLVMContext *ctx, Target target) {
LLVMTypes::VoidType = llvm::Type::getVoidTy(*ctx);
LLVMTypes::VoidPointerType = llvm::PointerType::get(llvm::Type::getInt8Ty(*ctx), 0);
LLVMTypes::PointerIntType = target.is32Bit ? llvm::Type::getInt32Ty(*ctx) :
llvm::Type::getInt64Ty(*ctx);
LLVMTypes::BoolType = llvm::Type::getInt1Ty(*ctx);
LLVMTypes::Int8Type = llvm::Type::getInt8Ty(*ctx);
LLVMTypes::Int16Type = llvm::Type::getInt16Ty(*ctx);
LLVMTypes::Int32Type = llvm::Type::getInt32Ty(*ctx);
LLVMTypes::Int64Type = llvm::Type::getInt64Ty(*ctx);
LLVMTypes::FloatType = llvm::Type::getFloatTy(*ctx);
LLVMTypes::DoubleType = llvm::Type::getDoubleTy(*ctx);
LLVMTypes::Int8PointerType = llvm::PointerType::get(LLVMTypes::Int8Type, 0);
LLVMTypes::Int16PointerType = llvm::PointerType::get(LLVMTypes::Int16Type, 0);
LLVMTypes::Int32PointerType = llvm::PointerType::get(LLVMTypes::Int32Type, 0);
LLVMTypes::Int64PointerType = llvm::PointerType::get(LLVMTypes::Int64Type, 0);
LLVMTypes::FloatPointerType = llvm::PointerType::get(LLVMTypes::FloatType, 0);
LLVMTypes::DoublePointerType = llvm::PointerType::get(LLVMTypes::DoubleType, 0);
if (target.maskBitCount == 1)
LLVMTypes::MaskType = LLVMTypes::BoolVectorType =
llvm::VectorType::get(llvm::Type::getInt1Ty(*ctx), target.vectorWidth);
else {
Assert(target.maskBitCount == 32);
LLVMTypes::MaskType = LLVMTypes::BoolVectorType =
llvm::VectorType::get(llvm::Type::getInt32Ty(*ctx), target.vectorWidth);
}
LLVMTypes::Int1VectorType =
llvm::VectorType::get(llvm::Type::getInt1Ty(*ctx), target.vectorWidth);
LLVMTypes::Int8VectorType =
llvm::VectorType::get(LLVMTypes::Int8Type, target.vectorWidth);
LLVMTypes::Int16VectorType =
llvm::VectorType::get(LLVMTypes::Int16Type, target.vectorWidth);
LLVMTypes::Int32VectorType =
llvm::VectorType::get(LLVMTypes::Int32Type, target.vectorWidth);
LLVMTypes::Int64VectorType =
llvm::VectorType::get(LLVMTypes::Int64Type, target.vectorWidth);
LLVMTypes::FloatVectorType =
llvm::VectorType::get(LLVMTypes::FloatType, target.vectorWidth);
LLVMTypes::DoubleVectorType =
llvm::VectorType::get(LLVMTypes::DoubleType, target.vectorWidth);
LLVMTypes::Int8VectorPointerType = llvm::PointerType::get(LLVMTypes::Int8VectorType, 0);
LLVMTypes::Int16VectorPointerType = llvm::PointerType::get(LLVMTypes::Int16VectorType, 0);
LLVMTypes::Int32VectorPointerType = llvm::PointerType::get(LLVMTypes::Int32VectorType, 0);
LLVMTypes::Int64VectorPointerType = llvm::PointerType::get(LLVMTypes::Int64VectorType, 0);
LLVMTypes::FloatVectorPointerType = llvm::PointerType::get(LLVMTypes::FloatVectorType, 0);
LLVMTypes::DoubleVectorPointerType = llvm::PointerType::get(LLVMTypes::DoubleVectorType, 0);
LLVMTypes::VoidPointerVectorType = g->target.is32Bit ? LLVMTypes::Int32VectorType :
LLVMTypes::Int64VectorType;
LLVMTrue = llvm::ConstantInt::getTrue(*ctx);
LLVMFalse = llvm::ConstantInt::getFalse(*ctx);
std::vector<llvm::Constant *> maskOnes;
llvm::Constant *onMask = NULL;
if (target.maskBitCount == 1)
onMask = llvm::ConstantInt::get(llvm::Type::getInt1Ty(*ctx), 1,
false /*unsigned*/); // 0x1
else
onMask = llvm::ConstantInt::get(llvm::Type::getInt32Ty(*ctx), -1,
true /*signed*/); // 0xffffffff
for (int i = 0; i < target.vectorWidth; ++i)
maskOnes.push_back(onMask);
LLVMMaskAllOn = llvm::ConstantVector::get(maskOnes);
std::vector<llvm::Constant *> maskZeros;
llvm::Constant *offMask = NULL;
if (target.maskBitCount == 1)
offMask = llvm::ConstantInt::get(llvm::Type::getInt1Ty(*ctx), 0,
true /*signed*/);
else
offMask = llvm::ConstantInt::get(llvm::Type::getInt32Ty(*ctx), 0,
true /*signed*/);
for (int i = 0; i < target.vectorWidth; ++i)
maskZeros.push_back(offMask);
LLVMMaskAllOff = llvm::ConstantVector::get(maskZeros);
}
llvm::ConstantInt *
LLVMInt8(int8_t ival) {
return llvm::ConstantInt::get(llvm::Type::getInt8Ty(*g->ctx), ival,
true /*signed*/);
}
llvm::ConstantInt *
LLVMUInt8(uint8_t ival) {
return llvm::ConstantInt::get(llvm::Type::getInt8Ty(*g->ctx), ival,
false /*unsigned*/);
}
llvm::ConstantInt *
LLVMInt16(int16_t ival) {
return llvm::ConstantInt::get(llvm::Type::getInt16Ty(*g->ctx), ival,
true /*signed*/);
}
llvm::ConstantInt *
LLVMUInt16(uint16_t ival) {
return llvm::ConstantInt::get(llvm::Type::getInt16Ty(*g->ctx), ival,
false /*unsigned*/);
}
llvm::ConstantInt *
LLVMInt32(int32_t ival) {
return llvm::ConstantInt::get(llvm::Type::getInt32Ty(*g->ctx), ival,
true /*signed*/);
}
llvm::ConstantInt *
LLVMUInt32(uint32_t ival) {
return llvm::ConstantInt::get(llvm::Type::getInt32Ty(*g->ctx), ival,
false /*unsigned*/);
}
llvm::ConstantInt *
LLVMInt64(int64_t ival) {
return llvm::ConstantInt::get(llvm::Type::getInt64Ty(*g->ctx), ival,
true /*signed*/);
}
llvm::ConstantInt *
LLVMUInt64(uint64_t ival) {
return llvm::ConstantInt::get(llvm::Type::getInt64Ty(*g->ctx), ival,
false /*unsigned*/);
}
llvm::Constant *
LLVMFloat(float fval) {
return llvm::ConstantFP::get(llvm::Type::getFloatTy(*g->ctx), fval);
}
llvm::Constant *
LLVMDouble(double dval) {
return llvm::ConstantFP::get(llvm::Type::getDoubleTy(*g->ctx), dval);
}
llvm::Constant *
LLVMInt8Vector(int8_t ival) {
llvm::Constant *v = LLVMInt8(ival);
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(v);
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMInt8Vector(const int8_t *ivec) {
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(LLVMInt8(ivec[i]));
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMUInt8Vector(uint8_t ival) {
llvm::Constant *v = LLVMUInt8(ival);
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(v);
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMUInt8Vector(const uint8_t *ivec) {
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(LLVMUInt8(ivec[i]));
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMInt16Vector(int16_t ival) {
llvm::Constant *v = LLVMInt16(ival);
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(v);
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMInt16Vector(const int16_t *ivec) {
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(LLVMInt16(ivec[i]));
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMUInt16Vector(uint16_t ival) {
llvm::Constant *v = LLVMUInt16(ival);
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(v);
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMUInt16Vector(const uint16_t *ivec) {
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(LLVMUInt16(ivec[i]));
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMInt32Vector(int32_t ival) {
llvm::Constant *v = LLVMInt32(ival);
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(v);
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMInt32Vector(const int32_t *ivec) {
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(LLVMInt32(ivec[i]));
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMUInt32Vector(uint32_t ival) {
llvm::Constant *v = LLVMUInt32(ival);
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(v);
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMUInt32Vector(const uint32_t *ivec) {
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(LLVMUInt32(ivec[i]));
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMFloatVector(float fval) {
llvm::Constant *v = LLVMFloat(fval);
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(v);
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMFloatVector(const float *fvec) {
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(LLVMFloat(fvec[i]));
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMDoubleVector(double dval) {
llvm::Constant *v = LLVMDouble(dval);
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(v);
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMDoubleVector(const double *dvec) {
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(LLVMDouble(dvec[i]));
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMInt64Vector(int64_t ival) {
llvm::Constant *v = LLVMInt64(ival);
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(v);
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMInt64Vector(const int64_t *ivec) {
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(LLVMInt64(ivec[i]));
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMUInt64Vector(uint64_t ival) {
llvm::Constant *v = LLVMUInt64(ival);
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(v);
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMUInt64Vector(const uint64_t *ivec) {
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(LLVMUInt64(ivec[i]));
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMBoolVector(bool b) {
llvm::Constant *v;
if (LLVMTypes::BoolVectorType == LLVMTypes::Int32VectorType)
v = llvm::ConstantInt::get(LLVMTypes::Int32Type, b ? 0xffffffff : 0,
false /*unsigned*/);
else {
Assert(LLVMTypes::BoolVectorType->getElementType() ==
llvm::Type::getInt1Ty(*g->ctx));
v = b ? LLVMTrue : LLVMFalse;
}
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i)
vals.push_back(v);
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMBoolVector(const bool *bvec) {
std::vector<llvm::Constant *> vals;
for (int i = 0; i < g->target.vectorWidth; ++i) {
llvm::Constant *v;
if (LLVMTypes::BoolVectorType == LLVMTypes::Int32VectorType)
v = llvm::ConstantInt::get(LLVMTypes::Int32Type, bvec[i] ? 0xffffffff : 0,
false /*unsigned*/);
else {
Assert(LLVMTypes::BoolVectorType->getElementType() ==
llvm::Type::getInt1Ty(*g->ctx));
v = bvec[i] ? LLVMTrue : LLVMFalse;
}
vals.push_back(v);
}
return llvm::ConstantVector::get(vals);
}
llvm::Constant *
LLVMIntAsType(int64_t val, llvm::Type *type) {
llvm::VectorType *vecType =
llvm::dyn_cast<llvm::VectorType>(type);
if (vecType != NULL) {
llvm::Constant *v = llvm::ConstantInt::get(vecType->getElementType(),
val, true /* signed */);
std::vector<llvm::Constant *> vals;
for (int i = 0; i < (int)vecType->getNumElements(); ++i)
vals.push_back(v);
return llvm::ConstantVector::get(vals);
}
else
return llvm::ConstantInt::get(type, val, true /* signed */);
}
llvm::Constant *
LLVMUIntAsType(uint64_t val, llvm::Type *type) {
llvm::VectorType *vecType =
llvm::dyn_cast<llvm::VectorType>(type);
if (vecType != NULL) {
llvm::Constant *v = llvm::ConstantInt::get(vecType->getElementType(),
val, false /* unsigned */);
std::vector<llvm::Constant *> vals;
for (int i = 0; i < (int)vecType->getNumElements(); ++i)
vals.push_back(v);
return llvm::ConstantVector::get(vals);
}
else
return llvm::ConstantInt::get(type, val, false /* unsigned */);
}
/** Conservative test to see if two llvm::Values are equal. There are
(potentially many) cases where the two values actually are equal but
this will return false. However, if it does return true, the two
vectors definitely are equal.
*/
static bool
lValuesAreEqual(llvm::Value *v0, llvm::Value *v1,
std::vector<llvm::PHINode *> &seenPhi0,
std::vector<llvm::PHINode *> &seenPhi1) {
// Thanks to the fact that LLVM hashes and returns the same pointer for
// constants (of all sorts, even constant expressions), this first test
// actually catches a lot of cases. LLVM's SSA form also helps a lot
// with this..
if (v0 == v1)
return true;
Assert(seenPhi0.size() == seenPhi1.size());
for (unsigned int i = 0; i < seenPhi0.size(); ++i)
if (v0 == seenPhi0[i] && v1 == seenPhi1[i])
return true;
llvm::BinaryOperator *bo0 = llvm::dyn_cast<llvm::BinaryOperator>(v0);
llvm::BinaryOperator *bo1 = llvm::dyn_cast<llvm::BinaryOperator>(v1);
if (bo0 != NULL && bo1 != NULL) {
if (bo0->getOpcode() != bo1->getOpcode())
return false;
return (lValuesAreEqual(bo0->getOperand(0), bo1->getOperand(0),
seenPhi0, seenPhi1) &&
lValuesAreEqual(bo0->getOperand(1), bo1->getOperand(1),
seenPhi0, seenPhi1));
}
llvm::CastInst *cast0 = llvm::dyn_cast<llvm::CastInst>(v0);
llvm::CastInst *cast1 = llvm::dyn_cast<llvm::CastInst>(v1);
if (cast0 != NULL && cast1 != NULL) {
if (cast0->getOpcode() != cast1->getOpcode())
return NULL;
return lValuesAreEqual(cast0->getOperand(0), cast1->getOperand(0),
seenPhi0, seenPhi1);
}
llvm::PHINode *phi0 = llvm::dyn_cast<llvm::PHINode>(v0);
llvm::PHINode *phi1 = llvm::dyn_cast<llvm::PHINode>(v1);
if (phi0 != NULL && phi1 != NULL) {
if (phi0->getNumIncomingValues() != phi1->getNumIncomingValues())
return false;
seenPhi0.push_back(phi0);
seenPhi1.push_back(phi1);
unsigned int numIncoming = phi0->getNumIncomingValues();
// Check all of the incoming values: if all of them are all equal,
// then we're good.
bool anyFailure = false;
for (unsigned int i = 0; i < numIncoming; ++i) {
// FIXME: should it be ok if the incoming blocks are different,
// where we just return faliure in this case?
Assert(phi0->getIncomingBlock(i) == phi1->getIncomingBlock(i));
if (!lValuesAreEqual(phi0->getIncomingValue(i),
phi1->getIncomingValue(i), seenPhi0, seenPhi1)) {
anyFailure = true;
break;
}
}
seenPhi0.pop_back();
seenPhi1.pop_back();
return !anyFailure;
}
return false;
}
/** Given an llvm::Value known to be an integer, return its value as
an int64_t.
*/
static int64_t
lGetIntValue(llvm::Value *offset) {
llvm::ConstantInt *intOffset = llvm::dyn_cast<llvm::ConstantInt>(offset);
Assert(intOffset && (intOffset->getBitWidth() == 32 ||
intOffset->getBitWidth() == 64));
return intOffset->getSExtValue();
}
void
LLVMFlattenInsertChain(llvm::InsertElementInst *ie, int vectorWidth,
llvm::Value **elements) {
for (int i = 0; i < vectorWidth; ++i)
elements[i] = NULL;
while (ie != NULL) {
int64_t iOffset = lGetIntValue(ie->getOperand(2));
Assert(iOffset >= 0 && iOffset < vectorWidth);
Assert(elements[iOffset] == NULL);
// Get the scalar value from this insert
elements[iOffset] = ie->getOperand(1);
// Do we have another insert?
llvm::Value *insertBase = ie->getOperand(0);
ie = llvm::dyn_cast<llvm::InsertElementInst>(insertBase);
if (ie == NULL) {
if (llvm::isa<llvm::UndefValue>(insertBase))
return;
// Get the value out of a constant vector if that's what we
// have
llvm::ConstantVector *cv =
llvm::dyn_cast<llvm::ConstantVector>(insertBase);
// FIXME: this assert is a little questionable; we probably
// shouldn't fail in this case but should just return an
// incomplete result. But there aren't currently any known
// cases where we have anything other than an undef value or a
// constant vector at the base, so if that ever does happen,
// it'd be nice to know what happend so that perhaps we can
// handle it.
// FIXME: Also, should we handle ConstantDataVectors with
// LLVM3.1? What about ConstantAggregateZero values??
Assert(cv != NULL);
Assert(iOffset < (int)cv->getNumOperands());
elements[iOffset] = cv->getOperand((int32_t)iOffset);
}
}
}
bool
LLVMExtractVectorInts(llvm::Value *v, int64_t ret[], int *nElts) {
// Make sure we do in fact have a vector of integer values here
llvm::VectorType *vt =
llvm::dyn_cast<llvm::VectorType>(v->getType());
Assert(vt != NULL);
Assert(llvm::isa<llvm::IntegerType>(vt->getElementType()));
*nElts = (int)vt->getNumElements();
if (llvm::isa<llvm::ConstantAggregateZero>(v)) {
for (int i = 0; i < (int)vt->getNumElements(); ++i)
ret[i] = 0;
return true;
}
// Deal with the fact that LLVM3.1 and previous versions have different
// representations for vectors of constant ints...
#ifdef LLVM_3_1svn
llvm::ConstantDataVector *cv = llvm::dyn_cast<llvm::ConstantDataVector>(v);
if (cv == NULL)
return false;
for (int i = 0; i < (int)cv->getNumElements(); ++i)
ret[i] = cv->getElementAsInteger(i);
return true;
#else
llvm::ConstantVector *cv = llvm::dyn_cast<llvm::ConstantVector>(v);
if (cv == NULL)
return false;
llvm::SmallVector<llvm::Constant *, ISPC_MAX_NVEC> elements;
cv->getVectorElements(elements);
for (int i = 0; i < (int)vt->getNumElements(); ++i) {
llvm::ConstantInt *ci = llvm::dyn_cast<llvm::ConstantInt>(elements[i]);
Assert(ci != NULL);
ret[i] = ci->getSExtValue();
}
return true;
#endif // LLVM_3_1svn
}
static bool
lVectorValuesAllEqual(llvm::Value *v, int vectorLength,
std::vector<llvm::PHINode *> &seenPhis);
/** This function checks to see if the given (scalar or vector) value is an
exact multiple of baseValue. It returns true if so, and false if not
(or if it's not able to determine if it is). Any vector value passed
in is required to have the same value in all elements (so that we can
just check the first element to be a multiple of the given value.)
*/
static bool
lIsExactMultiple(llvm::Value *val, int baseValue, int vectorLength,
std::vector<llvm::PHINode *> &seenPhis) {
if (llvm::isa<llvm::VectorType>(val->getType()) == false) {
// If we've worked down to a constant int, then the moment of truth
// has arrived...
llvm::ConstantInt *ci = llvm::dyn_cast<llvm::ConstantInt>(val);
if (ci != NULL)
return (ci->getZExtValue() % baseValue) == 0;
}
else
Assert(LLVMVectorValuesAllEqual(val));
llvm::InsertElementInst *ie = llvm::dyn_cast<llvm::InsertElementInst>(val);
if (ie != NULL) {
llvm::Value *elts[ISPC_MAX_NVEC];
LLVMFlattenInsertChain(ie, g->target.vectorWidth, elts);
// We just need to check the scalar first value, since we know that
// all elements are equal
return lIsExactMultiple(elts[0], baseValue, vectorLength,
seenPhis);
}
llvm::PHINode *phi = llvm::dyn_cast<llvm::PHINode>(val);
if (phi != NULL) {
for (unsigned int i = 0; i < seenPhis.size(); ++i)
if (phi == seenPhis[i])
return true;
seenPhis.push_back(phi);
unsigned int numIncoming = phi->getNumIncomingValues();
// Check all of the incoming values: if all of them pass, then
// we're good.
for (unsigned int i = 0; i < numIncoming; ++i) {
llvm::Value *incoming = phi->getIncomingValue(i);
bool mult = lIsExactMultiple(incoming, baseValue, vectorLength,
seenPhis);
if (mult == false) {
seenPhis.pop_back();
return false;
}
}
seenPhis.pop_back();
return true;
}
llvm::BinaryOperator *bop = llvm::dyn_cast<llvm::BinaryOperator>(val);
if (bop != NULL && bop->getOpcode() == llvm::Instruction::Add) {
llvm::Value *op0 = bop->getOperand(0);
llvm::Value *op1 = bop->getOperand(1);
bool be0 = lIsExactMultiple(op0, baseValue, vectorLength, seenPhis);
bool be1 = lIsExactMultiple(op1, baseValue, vectorLength, seenPhis);
return (be0 && be1);
}
// FIXME: mul? casts? ... ?
return false;
}
/** Returns the next power of two greater than or equal to the given
value. */
static int
lRoundUpPow2(int v) {
v--;
v |= v >> 1;
v |= v >> 2;
v |= v >> 4;
v |= v >> 8;
v |= v >> 16;
return v+1;
}
/** Try to determine if all of the elements of the given vector value have
the same value when divided by the given baseValue. The function
returns true if this can be determined to be the case, and false
otherwise. (This function may fail to identify some cases where it
does in fact have this property, but should never report a given value
as being a multiple if it isn't!)
*/
static bool
lAllDivBaseEqual(llvm::Value *val, int64_t baseValue, int vectorLength,
std::vector<llvm::PHINode *> &seenPhis,
bool &canAdd) {
Assert(llvm::isa<llvm::VectorType>(val->getType()));
// Make sure the base value is a positive power of 2
Assert(baseValue > 0 && (baseValue & (baseValue-1)) == 0);
// The easy case
if (lVectorValuesAllEqual(val, vectorLength, seenPhis))
return true;
int64_t vecVals[ISPC_MAX_NVEC];
int nElts;
if (llvm::isa<llvm::VectorType>(val->getType()) &&
LLVMExtractVectorInts(val, vecVals, &nElts)) {
// If we have a vector of compile-time constant integer values,
// then go ahead and check them directly..
int64_t firstDiv = vecVals[0] / baseValue;
for (int i = 1; i < nElts; ++i)
if ((vecVals[i] / baseValue) != firstDiv)
return false;
return true;
}
llvm::PHINode *phi = llvm::dyn_cast<llvm::PHINode>(val);
if (phi != NULL) {
for (unsigned int i = 0; i < seenPhis.size(); ++i)
if (phi == seenPhis[i])
return true;
seenPhis.push_back(phi);
unsigned int numIncoming = phi->getNumIncomingValues();
// Check all of the incoming values: if all of them pass, then
// we're good.
for (unsigned int i = 0; i < numIncoming; ++i) {
llvm::Value *incoming = phi->getIncomingValue(i);
bool ca = canAdd;
bool mult = lAllDivBaseEqual(incoming, baseValue, vectorLength,
seenPhis, ca);
if (mult == false) {
seenPhis.pop_back();
return false;
}
}
seenPhis.pop_back();
return true;
}
llvm::BinaryOperator *bop = llvm::dyn_cast<llvm::BinaryOperator>(val);
if (bop != NULL && bop->getOpcode() == llvm::Instruction::Add &&
canAdd == true) {
llvm::Value *op0 = bop->getOperand(0);
llvm::Value *op1 = bop->getOperand(1);
// Otherwise we're only going to worry about the following case,
// which comes up often when looping over SOA data:
// ashr %val, <constant shift>
// where %val = add %smear, <0,1,2,3...>
// and where the maximum of the <0,...> vector in the add is less than
// 1<<(constant shift),
// and where %smear is a smear of a value that is a multiple of
// baseValue.
int64_t addConstants[ISPC_MAX_NVEC];
if (LLVMExtractVectorInts(op1, addConstants, &nElts) == false)
return false;
Assert(nElts == vectorLength);
// Do all of them give the same value when divided by baseValue?
int64_t firstConstDiv = addConstants[0] / baseValue;
for (int i = 1; i < vectorLength; ++i)
if ((addConstants[i] / baseValue) != firstConstDiv)
return false;
if (lVectorValuesAllEqual(op0, vectorLength, seenPhis) == false)
return false;
// Note that canAdd is a reference parameter; setting this ensures
// that we don't allow multiple adds in other parts of the chain of
// dependent values from here.
canAdd = false;
// Now we need to figure out the required alignment (in numbers of
// elements of the underlying type being indexed) of the value to
// which these integer addConstant[] values are being added to. We
// know that we have addConstant[] values that all give the same
// value when divided by baseValue, but we may need a less-strict
// alignment than baseValue depending on the actual values.
//
// As an example, consider a case where the baseValue alignment is
// 16, but the addConstants here are <0,1,2,3>. In that case, the
// value to which addConstants is added to only needs to be a
// multiple of 4. Conversely, if addConstants are <4,5,6,7>, then
// we need a multiple of 8 to ensure that the final added result
// will still have the same value for all vector elements when
// divided by baseValue.
//
// All that said, here we need to find the maximum value of any of
// the addConstants[], mod baseValue. If we round that up to the
// next power of 2, we'll have a value that will be no greater than
// baseValue and sometimes less.
int maxMod = addConstants[0] % baseValue;
for (int i = 1; i < vectorLength; ++i)
maxMod = std::max(maxMod, int(addConstants[i] % baseValue));
int requiredAlignment = lRoundUpPow2(maxMod);
std::vector<llvm::PHINode *> seenPhisEEM;
return lIsExactMultiple(op0, requiredAlignment, vectorLength,
seenPhisEEM);
}
// TODO: could handle mul by a vector of equal constant integer values
// and the like here and adjust the 'baseValue' value when it evenly
// divides, but unclear if it's worthwhile...
return false;
}
/** Given a vector shift right of some value by some amount, try to
determine if all of the elements of the final result have the same
value (i.e. whether the high bits are all equal, disregarding the low
bits that are shifted out.) Returns true if so, and false otherwise.
*/
static bool
lVectorShiftRightAllEqual(llvm::Value *val, llvm::Value *shift,
int vectorLength) {
// Are we shifting all elements by a compile-time constant amount? If
// not, give up.
int64_t shiftAmount[ISPC_MAX_NVEC];
int nElts;
if (LLVMExtractVectorInts(shift, shiftAmount, &nElts) == false)
return false;
Assert(nElts == vectorLength);
// Is it the same amount for all elements?
for (int i = 0; i < vectorLength; ++i)
if (shiftAmount[i] != shiftAmount[0])
return false;
// Now see if the value divided by (1 << shift) can be determined to
// have the same value for all vector elements.
int pow2 = 1 << shiftAmount[0];
bool canAdd = true;
std::vector<llvm::PHINode *> seenPhis;
bool eq = lAllDivBaseEqual(val, pow2, vectorLength, seenPhis, canAdd);
#if 0
fprintf(stderr, "check all div base equal:\n");
LLVMDumpValue(shift);
LLVMDumpValue(val);
fprintf(stderr, "----> %s\n\n", eq ? "true" : "false");
#endif
return eq;
}
static bool
lVectorValuesAllEqual(llvm::Value *v, int vectorLength,
std::vector<llvm::PHINode *> &seenPhis) {
if (vectorLength == 1)
return true;
if (llvm::isa<llvm::ConstantAggregateZero>(v))
return true;
llvm::ConstantVector *cv = llvm::dyn_cast<llvm::ConstantVector>(v);
if (cv != NULL)
return (cv->getSplatValue() != NULL);
#ifdef LLVM_3_1svn
llvm::ConstantDataVector *cdv = llvm::dyn_cast<llvm::ConstantDataVector>(v);
if (cdv != NULL)
return (cdv->getSplatValue() != NULL);
#endif
llvm::BinaryOperator *bop = llvm::dyn_cast<llvm::BinaryOperator>(v);
if (bop != NULL) {
// Easy case: both operands are all equal -> return true
if (lVectorValuesAllEqual(bop->getOperand(0), vectorLength,
seenPhis) &&
lVectorValuesAllEqual(bop->getOperand(1), vectorLength,
seenPhis))
return true;
// If it's a shift, take a special path that tries to check if the
// high (surviving) bits of the values are equal.
if (bop->getOpcode() == llvm::Instruction::AShr ||
bop->getOpcode() == llvm::Instruction::LShr)
return lVectorShiftRightAllEqual(bop->getOperand(0),
bop->getOperand(1), vectorLength);
return false;
}
llvm::CastInst *cast = llvm::dyn_cast<llvm::CastInst>(v);
if (cast != NULL)
return lVectorValuesAllEqual(cast->getOperand(0), vectorLength,
seenPhis);
llvm::InsertElementInst *ie = llvm::dyn_cast<llvm::InsertElementInst>(v);
if (ie != NULL) {
llvm::Value *elements[ISPC_MAX_NVEC];
LLVMFlattenInsertChain(ie, vectorLength, elements);
// We will ignore any values of elements[] that are NULL; as they
// correspond to undefined values--we just want to see if all of
// the defined values have the same value.
int lastNonNull = 0;
while (lastNonNull < vectorLength && elements[lastNonNull] == NULL)
++lastNonNull;
if (lastNonNull == vectorLength)
// all of them are undef!
return true;
for (int i = lastNonNull; i < vectorLength; ++i) {
if (elements[i] == NULL)
continue;
std::vector<llvm::PHINode *> seenPhi0;
std::vector<llvm::PHINode *> seenPhi1;
if (lValuesAreEqual(elements[lastNonNull], elements[i], seenPhi0,
seenPhi1) == false)
return false;
lastNonNull = i;
}
return true;
}
llvm::PHINode *phi = llvm::dyn_cast<llvm::PHINode>(v);
if (phi) {
for (unsigned int i = 0; i < seenPhis.size(); ++i)
if (seenPhis[i] == phi)
return true;
seenPhis.push_back(phi);
unsigned int numIncoming = phi->getNumIncomingValues();
// Check all of the incoming values: if all of them are all equal,
// then we're good.
for (unsigned int i = 0; i < numIncoming; ++i) {
if (!lVectorValuesAllEqual(phi->getIncomingValue(i), vectorLength,
seenPhis)) {
seenPhis.pop_back();
return false;
}
}
seenPhis.pop_back();
return true;
}
if (llvm::isa<llvm::UndefValue>(v))
// ?
return false;
Assert(!llvm::isa<llvm::Constant>(v));
if (llvm::isa<llvm::CallInst>(v) || llvm::isa<llvm::LoadInst>(v) ||
!llvm::isa<llvm::Instruction>(v))
return false;
llvm::ShuffleVectorInst *shuffle = llvm::dyn_cast<llvm::ShuffleVectorInst>(v);
if (shuffle != NULL) {
llvm::Value *indices = shuffle->getOperand(2);
if (lVectorValuesAllEqual(indices, vectorLength, seenPhis))
// The easy case--just a smear of the same element across the
// whole vector.
return true;
// TODO: handle more general cases?
return false;
}
#if 0
fprintf(stderr, "all equal: ");
v->dump();
fprintf(stderr, "\n");
llvm::Instruction *inst = llvm::dyn_cast<llvm::Instruction>(v);
if (inst) {
inst->getParent()->dump();
fprintf(stderr, "\n");
fprintf(stderr, "\n");
}
#endif
return false;
}
/** Tests to see if all of the elements of the vector in the 'v' parameter
are equal. This is a conservative test and may return false for arrays
where the values are actually all equal.
*/
bool
LLVMVectorValuesAllEqual(llvm::Value *v) {
llvm::VectorType *vt =
llvm::dyn_cast<llvm::VectorType>(v->getType());
Assert(vt != NULL);
int vectorLength = vt->getNumElements();
std::vector<llvm::PHINode *> seenPhis;
bool equal = lVectorValuesAllEqual(v, vectorLength, seenPhis);
Debug(SourcePos(), "LLVMVectorValuesAllEqual(%s) -> %s.",
v->getName().str().c_str(), equal ? "true" : "false");
if (g->debugPrint)
LLVMDumpValue(v);
return equal;
}
static bool
lVectorIsLinear(llvm::Value *v, int vectorLength, int stride,
std::vector<llvm::PHINode *> &seenPhis);
/** Given a vector of compile-time constant integer values, test to see if
they are a linear sequence of constant integers starting from an
arbirary value but then having a step of value "stride" between
elements.
*/
static bool
lVectorIsLinearConstantInts(
#ifdef LLVM_3_1svn
llvm::ConstantDataVector *cv,
#else
llvm::ConstantVector *cv,
#endif
int vectorLength,
int stride) {
// Flatten the vector out into the elements array
llvm::SmallVector<llvm::Constant *, ISPC_MAX_NVEC> elements;
#ifdef LLVM_3_1svn
for (int i = 0; i < (int)cv->getNumElements(); ++i)
elements.push_back(cv->getElementAsConstant(i));
#else
cv->getVectorElements(elements);
#endif
Assert((int)elements.size() == vectorLength);
llvm::ConstantInt *ci = llvm::dyn_cast<llvm::ConstantInt>(elements[0]);
if (ci == NULL)
// Not a vector of integers
return false;
int64_t prevVal = ci->getSExtValue();
// For each element in the array, see if it is both a ConstantInt and
// if the difference between it and the value of the previous element
// is stride. If not, fail.
for (int i = 1; i < vectorLength; ++i) {
ci = llvm::dyn_cast<llvm::ConstantInt>(elements[i]);
if (ci == NULL)
return false;
int64_t nextVal = ci->getSExtValue();
if (prevVal + stride != nextVal)
return false;
prevVal = nextVal;
}
return true;
}
/** Checks to see if (op0 * op1) is a linear vector where the result is a
vector with values that increase by stride.
*/
static bool
lCheckMulForLinear(llvm::Value *op0, llvm::Value *op1, int vectorLength,
int stride, std::vector<llvm::PHINode *> &seenPhis) {
// Is the first operand a constant integer value splatted across all of
// the lanes?
#ifdef LLVM_3_1svn
llvm::ConstantDataVector *cv = llvm::dyn_cast<llvm::ConstantDataVector>(op0);
#else
llvm::ConstantVector *cv = llvm::dyn_cast<llvm::ConstantVector>(op0);
#endif
if (cv == NULL)
return false;
llvm::Constant *csplat = cv->getSplatValue();
if (csplat == NULL)
return false;
llvm::ConstantInt *splat = llvm::dyn_cast<llvm::ConstantInt>(csplat);
if (splat == NULL)
return false;
// If the splat value doesn't evenly divide the stride we're looking
// for, there's no way that we can get the linear sequence we're
// looking or.
int64_t splatVal = splat->getSExtValue();
if (splatVal == 0 || splatVal > stride || (stride % splatVal) != 0)
return false;
// Check to see if the other operand is a linear vector with stride
// given by stride/splatVal.
return lVectorIsLinear(op1, vectorLength, (int)(stride / splatVal),
seenPhis);
}
/** Given (op0 AND op1), try and see if we can determine if the result is a
linear sequence with a step of "stride" between values. Returns true
if so and false otherwise. This pattern comes up when accessing SOA
data.
*/
static bool
lCheckAndForLinear(llvm::Value *op0, llvm::Value *op1, int vectorLength,
int stride, std::vector<llvm::PHINode *> &seenPhis) {
// Require op1 to be a compile-time constant
int64_t maskValue[ISPC_MAX_NVEC];
int nElts;
if (LLVMExtractVectorInts(op1, maskValue, &nElts) == false)
return false;
Assert(nElts == vectorLength);
// Is op1 a smear of the same value across all lanes? Give up if not.
for (int i = 1; i < vectorLength; ++i)
if (maskValue[i] != maskValue[0])
return false;
// If the op1 value isn't power of 2 minus one, then also give up.
int64_t maskPlusOne = maskValue[0] + 1;
bool isPowTwo = (maskPlusOne & (maskPlusOne - 1)) == 0;
if (isPowTwo == false)
return false;
// The case we'll covert here is op0 being a linear vector with desired
// stride, and where all of the values of op0, when divided by
// maskPlusOne, have the same value.
if (lVectorIsLinear(op0, vectorLength, stride, seenPhis) == false)
return false;
bool canAdd = true;
bool isMult = lAllDivBaseEqual(op0, maskPlusOne, vectorLength, seenPhis,
canAdd);
return isMult;
}
static bool
lVectorIsLinear(llvm::Value *v, int vectorLength, int stride,
std::vector<llvm::PHINode *> &seenPhis) {
// First try the easy case: if the values are all just constant
// integers and have the expected stride between them, then we're done.
#ifdef LLVM_3_1svn
llvm::ConstantDataVector *cv = llvm::dyn_cast<llvm::ConstantDataVector>(v);
#else
llvm::ConstantVector *cv = llvm::dyn_cast<llvm::ConstantVector>(v);
#endif
if (cv != NULL)
return lVectorIsLinearConstantInts(cv, vectorLength, stride);
llvm::BinaryOperator *bop = llvm::dyn_cast<llvm::BinaryOperator>(v);
if (bop != NULL) {
// FIXME: is it right to pass the seenPhis to the all equal check as well??
llvm::Value *op0 = bop->getOperand(0), *op1 = bop->getOperand(1);
if (bop->getOpcode() == llvm::Instruction::Add) {
// There are two cases to check if we have an add:
//
// programIndex + unif -> ascending linear seqeuence
// unif + programIndex -> ascending linear sequence
bool l0 = lVectorIsLinear(op0, vectorLength, stride, seenPhis);
bool e1 = lVectorValuesAllEqual(op1, vectorLength, seenPhis);
if (l0 && e1)
return true;
bool e0 = lVectorValuesAllEqual(op0, vectorLength, seenPhis);
bool l1 = lVectorIsLinear(op1, vectorLength, stride, seenPhis);
return (e0 && l1);
}
else if (bop->getOpcode() == llvm::Instruction::Sub)
// For subtraction, we only match:
// programIndex - unif -> ascending linear seqeuence
return (lVectorIsLinear(bop->getOperand(0), vectorLength,
stride, seenPhis) &&
lVectorValuesAllEqual(bop->getOperand(1), vectorLength,
seenPhis));
else if (bop->getOpcode() == llvm::Instruction::Mul) {
// Multiplies are a bit trickier, so are handled in a separate
// function.
bool m0 = lCheckMulForLinear(op0, op1, vectorLength, stride, seenPhis);
if (m0)
return true;
bool m1 = lCheckMulForLinear(op1, op0, vectorLength, stride, seenPhis);
return m1;
}
else if (bop->getOpcode() == llvm::Instruction::And) {
// Special case for some AND-related patterns that come up when
// looping over SOA data
bool linear = lCheckAndForLinear(op0, op1, vectorLength, stride, seenPhis);
return linear;
}
else
return false;
}
llvm::CastInst *ci = llvm::dyn_cast<llvm::CastInst>(v);
if (ci != NULL)
return lVectorIsLinear(ci->getOperand(0), vectorLength,
stride, seenPhis);
if (llvm::isa<llvm::CallInst>(v) || llvm::isa<llvm::LoadInst>(v))
return false;
llvm::PHINode *phi = llvm::dyn_cast<llvm::PHINode>(v);
if (phi != NULL) {
for (unsigned int i = 0; i < seenPhis.size(); ++i)
if (seenPhis[i] == phi)
return true;
seenPhis.push_back(phi);
unsigned int numIncoming = phi->getNumIncomingValues();
// Check all of the incoming values: if all of them are all equal,
// then we're good.
for (unsigned int i = 0; i < numIncoming; ++i) {
if (!lVectorIsLinear(phi->getIncomingValue(i), vectorLength, stride,
seenPhis)) {
seenPhis.pop_back();
return false;
}
}
seenPhis.pop_back();
return true;
}
// TODO: is any reason to worry about these?
if (llvm::isa<llvm::InsertElementInst>(v))
return false;
// TODO: we could also handle shuffles, but we haven't yet seen any
// cases where doing so would detect cases where actually have a linear
// vector.
llvm::ShuffleVectorInst *shuffle = llvm::dyn_cast<llvm::ShuffleVectorInst>(v);
if (shuffle != NULL)
return false;
#if 0
fprintf(stderr, "linear check: ");
v->dump();
fprintf(stderr, "\n");
llvm::Instruction *inst = llvm::dyn_cast<llvm::Instruction>(v);
if (inst) {
inst->getParent()->dump();
fprintf(stderr, "\n");
fprintf(stderr, "\n");
}
#endif
return false;
}
/** Given vector of integer-typed values, see if the elements of the array
have a step of 'stride' between their values. This function tries to
handle as many possibilities as possible, including things like all
elements equal to some non-constant value plus an integer offset, etc.
*/
bool
LLVMVectorIsLinear(llvm::Value *v, int stride) {
llvm::VectorType *vt =
llvm::dyn_cast<llvm::VectorType>(v->getType());
Assert(vt != NULL);
int vectorLength = vt->getNumElements();
std::vector<llvm::PHINode *> seenPhis;
bool linear = lVectorIsLinear(v, vectorLength, stride, seenPhis);
Debug(SourcePos(), "LLVMVectorIsLinear(%s) -> %s.",
v->getName().str().c_str(), linear ? "true" : "false");
if (g->debugPrint)
LLVMDumpValue(v);
return linear;
}
static void
lDumpValue(llvm::Value *v, std::set<llvm::Value *> &done) {
if (done.find(v) != done.end())
return;
llvm::Instruction *inst = llvm::dyn_cast<llvm::Instruction>(v);
if (done.size() > 0 && inst == NULL)
return;
fprintf(stderr, " ");
v->dump();
done.insert(v);
if (inst == NULL)
return;
for (unsigned i = 0; i < inst->getNumOperands(); ++i)
lDumpValue(inst->getOperand(i), done);
}
void
LLVMDumpValue(llvm::Value *v) {
std::set<llvm::Value *> done;
lDumpValue(v, done);
fprintf(stderr, "----\n");
}
static llvm::Value *
lExtractFirstVectorElement(llvm::Value *v,
std::map<llvm::PHINode *, llvm::PHINode *> &phiMap) {
llvm::VectorType *vt =
llvm::dyn_cast<llvm::VectorType>(v->getType());
Assert(vt != NULL);
// First, handle various constant types; do the extraction manually, as
// appropriate.
if (llvm::isa<llvm::ConstantAggregateZero>(v) == true) {
Assert(vt->getElementType()->isIntegerTy());
return llvm::ConstantInt::get(vt->getElementType(), 0);
}
if (llvm::ConstantVector *cv = llvm::dyn_cast<llvm::ConstantVector>(v)) {
#ifdef LLVM_3_1svn
return cv->getOperand(0);
#else
llvm::SmallVector<llvm::Constant *, ISPC_MAX_NVEC> elements;
cv->getVectorElements(elements);
return elements[0];
#endif // LLVM_3_1
}
#ifdef LLVM_3_1svn
if (llvm::ConstantDataVector *cdv =
llvm::dyn_cast<llvm::ConstantDataVector>(v))
return cdv->getElementAsConstant(0);
#endif // LLVM_3_1
// Otherwise, all that we should have at this point is an instruction
// of some sort
Assert(llvm::isa<llvm::Constant>(v) == false);
Assert(llvm::isa<llvm::Instruction>(v) == true);
std::string newName = v->getName().str() + std::string(".elt0");
// Rewrite regular binary operators and casts to the scalarized
// equivalent.
llvm::BinaryOperator *bop = llvm::dyn_cast<llvm::BinaryOperator>(v);
if (bop != NULL) {
llvm::Value *v0 = lExtractFirstVectorElement(bop->getOperand(0),
phiMap);
llvm::Value *v1 = lExtractFirstVectorElement(bop->getOperand(1),
phiMap);
// Note that the new binary operator is inserted immediately before
// the previous vector one
return llvm::BinaryOperator::Create(bop->getOpcode(), v0, v1,
newName, bop);
}
llvm::CastInst *cast = llvm::dyn_cast<llvm::CastInst>(v);
if (cast != NULL) {
llvm::Value *v = lExtractFirstVectorElement(cast->getOperand(0),
phiMap);
// Similarly, the equivalent scalar cast instruction goes right
// before the vector cast
return llvm::CastInst::Create(cast->getOpcode(), v,
vt->getElementType(), newName,
cast);
}
llvm::PHINode *phi = llvm::dyn_cast<llvm::PHINode>(v);
if (phi != NULL) {
// For PHI notes, recursively scalarize them.
if (phiMap.find(phi) != phiMap.end())
return phiMap[phi];
// We need to create the new scalar PHI node immediately, though,
// and put it in the map<>, so that if we come back to this node
// via a recursive lExtractFirstVectorElement() call, then we can
// return the pointer and not get stuck in an infinite loop.
//
// The insertion point for the new phi node also has to be the
// start of the bblock of the original phi node.
llvm::Instruction *phiInsertPos = phi->getParent()->begin();
llvm::PHINode *scalarPhi =
llvm::PHINode::Create(vt->getElementType(),
phi->getNumIncomingValues(),
newName, phiInsertPos);
phiMap[phi] = scalarPhi;
for (unsigned i = 0; i < phi->getNumIncomingValues(); ++i) {
llvm::Value *v = lExtractFirstVectorElement(phi->getIncomingValue(i),
phiMap);
scalarPhi->addIncoming(v, phi->getIncomingBlock(i));
}
return scalarPhi;
}
// If we have a chain of insertelement instructions, then we can just
// flatten them out and grab the value for the first one.
llvm::InsertElementInst *ie = llvm::dyn_cast<llvm::InsertElementInst>(v);
if (ie != NULL) {
llvm::Value *elements[ISPC_MAX_NVEC];
LLVMFlattenInsertChain(ie, vt->getNumElements(), elements);
return elements[0];
}
// Worst case, for everything else, just do a regular extract element
// instruction, which we insert immediately after the instruction we
// have here.
llvm::Instruction *insertAfter = llvm::dyn_cast<llvm::Instruction>(v);
Assert(insertAfter != NULL);
llvm::Instruction *ee =
llvm::ExtractElementInst::Create(v, LLVMInt32(0), "first_elt",
(llvm::Instruction *)NULL);
ee->insertAfter(insertAfter);
return ee;
}
llvm::Value *
LLVMExtractFirstVectorElement(llvm::Value *v) {
std::map<llvm::PHINode *, llvm::PHINode *> phiMap;
llvm::Value *ret = lExtractFirstVectorElement(v, phiMap);
return ret;
}
/** Given two vectors of the same type, concatenate them into a vector that
has twice as many elements, where the first half has the elements from
the first vector and the second half has the elements from the second
vector.
*/
llvm::Value *
LLVMConcatVectors(llvm::Value *v1, llvm::Value *v2,
llvm::Instruction *insertBefore) {
Assert(v1->getType() == v2->getType());
llvm::VectorType *vt =
llvm::dyn_cast<llvm::VectorType>(v1->getType());
Assert(vt != NULL);
int32_t identity[ISPC_MAX_NVEC];
int resultSize = 2*vt->getNumElements();
Assert(resultSize <= ISPC_MAX_NVEC);
for (int i = 0; i < resultSize; ++i)
identity[i] = i;
return LLVMShuffleVectors(v1, v2, identity, resultSize, insertBefore);
}
/** Shuffle two vectors together with a ShuffleVectorInst, returning a
vector with shufSize elements, where the shuf[] array offsets are used
to determine which element from the two given vectors is used for each
result element. */
llvm::Value *
LLVMShuffleVectors(llvm::Value *v1, llvm::Value *v2, int32_t shuf[],
int shufSize, llvm::Instruction *insertBefore) {
std::vector<llvm::Constant *> shufVec;
for (int i = 0; i < shufSize; ++i) {
if (shuf[i] == -1)
shufVec.push_back(llvm::UndefValue::get(LLVMTypes::Int32Type));
else
shufVec.push_back(LLVMInt32(shuf[i]));
}
llvm::ArrayRef<llvm::Constant *> aref(&shufVec[0], &shufVec[shufSize]);
llvm::Value *vec = llvm::ConstantVector::get(aref);
return new llvm::ShuffleVectorInst(v1, v2, vec, "shuffle", insertBefore);
}
Reference in New Issue View Git Blame Copy Permalink