RPCSX/llvm
1
0
Fork 0
mirror of https://github.com/RPCSX/llvm.git synced 2025-01-23 20:57:21 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

[LV] Move addRuntimeCheck to LoopAccessAnalysis

Browse Source 
This will allow it to be shared with the new Loop Distribution pass.

getFirstInst is currently duplicated across LoopVectorize.cpp and
LoopAccessAnalysis.cpp.  This is a short-term work-around until we figure out
a better solution.

NFC.  (The code moved is adjusted a bit for the name of the Loop member and
that PtrRtCheck is now a reference rather than a pointer.)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228418 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Adam Nemet 2015-02-06 18:31:04 +00:00
parent 6dc42dd2da
commit b3189eac3f
3 changed files with 118 additions and 104 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
include/llvm/Analysis/LoopAccessAnalysis.h
View File

@ -163,6 +163,13 @@ public:
unsigned getNumStores() const { return NumStores; }
unsigned getNumLoads() const { return NumLoads;}
/// \brief Add code that checks at runtime if the accessed arrays overlap.
///
/// Returns a pair of instructions where the first element is the first
/// instruction generated in possibly a sequence of instructions and the
/// second value is the final comparator value or NULL if no check is needed.
std::pair<Instruction *, Instruction *> addRuntimeCheck(Instruction *Loc);
private:
void emitAnalysis(VectorizationReport &Message);

106
lib/Analysis/LoopAccessAnalysis.cpp
View File

@ -14,9 +14,11 @@
#include "llvm/Analysis/LoopAccessAnalysis.h"
#include "llvm/Analysis/LoopInfo.h"
#include "llvm/Analysis/ScalarEvolutionExpander.h"
#include "llvm/Analysis/ValueTracking.h"
#include "llvm/IR/DiagnosticInfo.h"
#include "llvm/IR/Dominators.h"
#include "llvm/IR/IRBuilder.h"
#include "llvm/Support/Debug.h"
#include "llvm/Transforms/Utils/VectorUtils.h"
using namespace llvm;
@ -1082,3 +1084,107 @@ void LoopAccessAnalysis::emitAnalysis(VectorizationReport &Message) {
bool LoopAccessAnalysis::isUniform(Value *V) {
return (SE->isLoopInvariant(SE->getSCEV(V), TheLoop));
}
// FIXME: this function is currently a duplicate of the one in
// LoopVectorize.cpp.
static Instruction *getFirstInst(Instruction *FirstInst, Value *V,
Instruction *Loc) {
if (FirstInst)
return FirstInst;
if (Instruction *I = dyn_cast<Instruction>(V))
return I->getParent() == Loc->getParent() ? I : nullptr;
return nullptr;
}
std::pair<Instruction *, Instruction *>
LoopAccessAnalysis::addRuntimeCheck(Instruction *Loc) {
Instruction *tnullptr = nullptr;
if (!PtrRtCheck.Need)
return std::pair<Instruction *, Instruction *>(tnullptr, tnullptr);
unsigned NumPointers = PtrRtCheck.Pointers.size();
SmallVector<TrackingVH<Value> , 2> Starts;
SmallVector<TrackingVH<Value> , 2> Ends;
LLVMContext &Ctx = Loc->getContext();
SCEVExpander Exp(*SE, "induction");
Instruction *FirstInst = nullptr;
for (unsigned i = 0; i < NumPointers; ++i) {
Value *Ptr = PtrRtCheck.Pointers[i];
const SCEV *Sc = SE->getSCEV(Ptr);
if (SE->isLoopInvariant(Sc, TheLoop)) {
DEBUG(dbgs() << "LV: Adding RT check for a loop invariant ptr:" <<
*Ptr <<"\n");
Starts.push_back(Ptr);
Ends.push_back(Ptr);
} else {
DEBUG(dbgs() << "LV: Adding RT check for range:" << *Ptr << '\n');
unsigned AS = Ptr->getType()->getPointerAddressSpace();
// Use this type for pointer arithmetic.
Type *PtrArithTy = Type::getInt8PtrTy(Ctx, AS);
Value *Start = Exp.expandCodeFor(PtrRtCheck.Starts[i], PtrArithTy, Loc);
Value *End = Exp.expandCodeFor(PtrRtCheck.Ends[i], PtrArithTy, Loc);
Starts.push_back(Start);
Ends.push_back(End);
}
}
IRBuilder<> ChkBuilder(Loc);
// Our instructions might fold to a constant.
Value *MemoryRuntimeCheck = nullptr;
for (unsigned i = 0; i < NumPointers; ++i) {
for (unsigned j = i+1; j < NumPointers; ++j) {
// No need to check if two readonly pointers intersect.
if (!PtrRtCheck.IsWritePtr[i] && !PtrRtCheck.IsWritePtr[j])
continue;
// Only need to check pointers between two different dependency sets.
if (PtrRtCheck.DependencySetId[i] == PtrRtCheck.DependencySetId[j])
continue;
// Only need to check pointers in the same alias set.
if (PtrRtCheck.AliasSetId[i] != PtrRtCheck.AliasSetId[j])
continue;
unsigned AS0 = Starts[i]->getType()->getPointerAddressSpace();
unsigned AS1 = Starts[j]->getType()->getPointerAddressSpace();
assert((AS0 == Ends[j]->getType()->getPointerAddressSpace()) &&
(AS1 == Ends[i]->getType()->getPointerAddressSpace()) &&
"Trying to bounds check pointers with different address spaces");
Type *PtrArithTy0 = Type::getInt8PtrTy(Ctx, AS0);
Type *PtrArithTy1 = Type::getInt8PtrTy(Ctx, AS1);
Value *Start0 = ChkBuilder.CreateBitCast(Starts[i], PtrArithTy0, "bc");
Value *Start1 = ChkBuilder.CreateBitCast(Starts[j], PtrArithTy1, "bc");
Value *End0 = ChkBuilder.CreateBitCast(Ends[i], PtrArithTy1, "bc");
Value *End1 = ChkBuilder.CreateBitCast(Ends[j], PtrArithTy0, "bc");
Value *Cmp0 = ChkBuilder.CreateICmpULE(Start0, End1, "bound0");
FirstInst = getFirstInst(FirstInst, Cmp0, Loc);
Value *Cmp1 = ChkBuilder.CreateICmpULE(Start1, End0, "bound1");
FirstInst = getFirstInst(FirstInst, Cmp1, Loc);
Value *IsConflict = ChkBuilder.CreateAnd(Cmp0, Cmp1, "found.conflict");
FirstInst = getFirstInst(FirstInst, IsConflict, Loc);
if (MemoryRuntimeCheck) {
IsConflict = ChkBuilder.CreateOr(MemoryRuntimeCheck, IsConflict,
"conflict.rdx");
FirstInst = getFirstInst(FirstInst, IsConflict, Loc);
}
MemoryRuntimeCheck = IsConflict;
}
}
// We have to do this trickery because the IRBuilder might fold the check to a
// constant expression in which case there is no Instruction anchored in a
// the block.
Instruction *Check = BinaryOperator::CreateAnd(MemoryRuntimeCheck,
ConstantInt::getTrue(Ctx));
ChkBuilder.Insert(Check, "memcheck.conflict");
FirstInst = getFirstInst(FirstInst, Check, Loc);
return std::make_pair(FirstInst, Check);
}

109
lib/Transforms/Vectorize/LoopVectorize.cpp
View File

@ -271,13 +271,6 @@ protected:
typedef DenseMap<std::pair<BasicBlock*, BasicBlock*>,
VectorParts> EdgeMaskCache;
/// \brief Add code that checks at runtime if the accessed arrays overlap.
///
/// Returns a pair of instructions where the first element is the first
/// instruction generated in possibly a sequence of instructions and the
/// second value is the final comparator value or NULL if no check is needed.
std::pair<Instruction *, Instruction *> addRuntimeCheck(Instruction *Loc);
/// \brief Add checks for strides that where assumed to be 1.
///
/// Returns the last check instruction and the first check instruction in the
@ -751,6 +744,10 @@ public:
return LAA.getRuntimePointerCheck();
}
LoopAccessAnalysis *getLAA() {
return &LAA;
}
/// This function returns the identity element (or neutral element) for
/// the operation K.
static Constant *getReductionIdentity(ReductionKind K, Type *Tp);
@ -2009,102 +2006,6 @@ InnerLoopVectorizer::addStrideCheck(Instruction *Loc) {
return std::make_pair(FirstInst, TheCheck);
}
std::pair<Instruction *, Instruction *>
InnerLoopVectorizer::addRuntimeCheck(Instruction *Loc) {
LoopAccessAnalysis::RuntimePointerCheck *PtrRtCheck =
Legal->getRuntimePointerCheck();
Instruction *tnullptr = nullptr;
if (!PtrRtCheck->Need)
return std::pair<Instruction *, Instruction *>(tnullptr, tnullptr);
unsigned NumPointers = PtrRtCheck->Pointers.size();
SmallVector<TrackingVH<Value> , 2> Starts;
SmallVector<TrackingVH<Value> , 2> Ends;
LLVMContext &Ctx = Loc->getContext();
SCEVExpander Exp(*SE, "induction");
Instruction *FirstInst = nullptr;
for (unsigned i = 0; i < NumPointers; ++i) {
Value *Ptr = PtrRtCheck->Pointers[i];
const SCEV *Sc = SE->getSCEV(Ptr);
if (SE->isLoopInvariant(Sc, OrigLoop)) {
DEBUG(dbgs() << "LV: Adding RT check for a loop invariant ptr:" <<
*Ptr <<"\n");
Starts.push_back(Ptr);
Ends.push_back(Ptr);
} else {
DEBUG(dbgs() << "LV: Adding RT check for range:" << *Ptr << '\n');
unsigned AS = Ptr->getType()->getPointerAddressSpace();
// Use this type for pointer arithmetic.
Type *PtrArithTy = Type::getInt8PtrTy(Ctx, AS);
Value *Start = Exp.expandCodeFor(PtrRtCheck->Starts[i], PtrArithTy, Loc);
Value *End = Exp.expandCodeFor(PtrRtCheck->Ends[i], PtrArithTy, Loc);
Starts.push_back(Start);
Ends.push_back(End);
}
}
IRBuilder<> ChkBuilder(Loc);
// Our instructions might fold to a constant.
Value *MemoryRuntimeCheck = nullptr;
for (unsigned i = 0; i < NumPointers; ++i) {
for (unsigned j = i+1; j < NumPointers; ++j) {
// No need to check if two readonly pointers intersect.
if (!PtrRtCheck->IsWritePtr[i] && !PtrRtCheck->IsWritePtr[j])
continue;
// Only need to check pointers between two different dependency sets.
if (PtrRtCheck->DependencySetId[i] == PtrRtCheck->DependencySetId[j])
continue;
// Only need to check pointers in the same alias set.
if (PtrRtCheck->AliasSetId[i] != PtrRtCheck->AliasSetId[j])
continue;
unsigned AS0 = Starts[i]->getType()->getPointerAddressSpace();
unsigned AS1 = Starts[j]->getType()->getPointerAddressSpace();
assert((AS0 == Ends[j]->getType()->getPointerAddressSpace()) &&
(AS1 == Ends[i]->getType()->getPointerAddressSpace()) &&
"Trying to bounds check pointers with different address spaces");
Type *PtrArithTy0 = Type::getInt8PtrTy(Ctx, AS0);
Type *PtrArithTy1 = Type::getInt8PtrTy(Ctx, AS1);
Value *Start0 = ChkBuilder.CreateBitCast(Starts[i], PtrArithTy0, "bc");
Value *Start1 = ChkBuilder.CreateBitCast(Starts[j], PtrArithTy1, "bc");
Value *End0 = ChkBuilder.CreateBitCast(Ends[i], PtrArithTy1, "bc");
Value *End1 = ChkBuilder.CreateBitCast(Ends[j], PtrArithTy0, "bc");
Value *Cmp0 = ChkBuilder.CreateICmpULE(Start0, End1, "bound0");
FirstInst = getFirstInst(FirstInst, Cmp0, Loc);
Value *Cmp1 = ChkBuilder.CreateICmpULE(Start1, End0, "bound1");
FirstInst = getFirstInst(FirstInst, Cmp1, Loc);
Value *IsConflict = ChkBuilder.CreateAnd(Cmp0, Cmp1, "found.conflict");
FirstInst = getFirstInst(FirstInst, IsConflict, Loc);
if (MemoryRuntimeCheck) {
IsConflict = ChkBuilder.CreateOr(MemoryRuntimeCheck, IsConflict,
"conflict.rdx");
FirstInst = getFirstInst(FirstInst, IsConflict, Loc);
}
MemoryRuntimeCheck = IsConflict;
}
}
// We have to do this trickery because the IRBuilder might fold the check to a
// constant expression in which case there is no Instruction anchored in a
// the block.
Instruction *Check = BinaryOperator::CreateAnd(MemoryRuntimeCheck,
ConstantInt::getTrue(Ctx));
ChkBuilder.Insert(Check, "memcheck.conflict");
FirstInst = getFirstInst(FirstInst, Check, Loc);
return std::make_pair(FirstInst, Check);
}
void InnerLoopVectorizer::createEmptyLoop() {
/*
In this function we generate a new loop. The new loop will contain
@ -2329,7 +2230,7 @@ void InnerLoopVectorizer::createEmptyLoop() {
// faster.
Instruction *MemRuntimeCheck;
std::tie(FirstCheckInst, MemRuntimeCheck) =
addRuntimeCheck(LastBypassBlock->getTerminator());
Legal->getLAA()->addRuntimeCheck(LastBypassBlock->getTerminator());
if (MemRuntimeCheck) {
// Create a new block containing the memory check.
BasicBlock *CheckBlock =