#include "llvm/IR/GlobalVariable.h"
#include "llvm/IR/IntrinsicInst.h"
#include "llvm/IR/LLVMContext.h"
+#include "llvm/IR/Module.h"
#include "llvm/IR/Operator.h"
using namespace llvm;
///
static bool AreEquivalentAddressValues(const Value *A, const Value *B) {
// Test if the values are trivially equivalent.
- if (A == B) return true;
+ if (A == B)
+ return true;
// Test if the values come from identical arithmetic instructions.
// Use isIdenticalToWhenDefined instead of isIdenticalTo because
// this function is only used when one address use dominates the
// other, which means that they'll always either have the same
// value or one of them will have an undefined value.
- if (isa<BinaryOperator>(A) || isa<CastInst>(A) ||
- isa<PHINode>(A) || isa<GetElementPtrInst>(A))
+ if (isa<BinaryOperator>(A) || isa<CastInst>(A) || isa<PHINode>(A) ||
+ isa<GetElementPtrInst>(A))
if (const Instruction *BI = dyn_cast<Instruction>(B))
if (cast<Instruction>(A)->isIdenticalToWhenDefined(BI))
return true;
return false;
}
-/// isSafeToLoadUnconditionally - Return true if we know that executing a load
-/// from this value cannot trap. If it is not obviously safe to load from the
-/// specified pointer, we do a quick local scan of the basic block containing
-/// ScanFrom, to determine if the address is already accessed.
+/// \brief Check if executing a load of this pointer value cannot trap.
+///
+/// If it is not obviously safe to load from the specified pointer, we do
+/// a quick local scan of the basic block containing \c ScanFrom, to determine
+/// if the address is already accessed.
+///
+/// This uses the pointee type to determine how many bytes need to be safe to
+/// load from the pointer.
bool llvm::isSafeToLoadUnconditionally(Value *V, Instruction *ScanFrom,
- unsigned Align, const DataLayout *TD) {
+ unsigned Align) {
+ const DataLayout &DL = ScanFrom->getModule()->getDataLayout();
+
+ // Zero alignment means that the load has the ABI alignment for the target
+ if (Align == 0)
+ Align = DL.getABITypeAlignment(V->getType()->getPointerElementType());
+ assert(isPowerOf2_32(Align));
+
int64_t ByteOffset = 0;
Value *Base = V;
- Base = GetPointerBaseWithConstantOffset(V, ByteOffset, TD);
+ Base = GetPointerBaseWithConstantOffset(V, ByteOffset, DL);
if (ByteOffset < 0) // out of bounds
return false;
BaseType = AI->getAllocatedType();
BaseAlign = AI->getAlignment();
} else if (const GlobalVariable *GV = dyn_cast<GlobalVariable>(Base)) {
- // Global variables are safe to load from but their size cannot be
- // guaranteed if they are overridden.
+ // Global variables are not necessarily safe to load from if they are
+ // overridden. Their size may change or they may be weak and require a test
+ // to determine if they were in fact provided.
if (!GV->mayBeOverridden()) {
BaseType = GV->getType()->getElementType();
BaseAlign = GV->getAlignment();
}
}
+ PointerType *AddrTy = cast<PointerType>(V->getType());
+ uint64_t LoadSize = DL.getTypeStoreSize(AddrTy->getElementType());
+
+ // If we found a base allocated type from either an alloca or global variable,
+ // try to see if we are definitively within the allocated region. We need to
+ // know the size of the base type and the loaded type to do anything in this
+ // case.
if (BaseType && BaseType->isSized()) {
- if (TD && BaseAlign == 0)
- BaseAlign = TD->getPrefTypeAlignment(BaseType);
+ if (BaseAlign == 0)
+ BaseAlign = DL.getPrefTypeAlignment(BaseType);
if (Align <= BaseAlign) {
- if (!TD)
- return true; // Loading directly from an alloca or global is OK.
-
// Check if the load is within the bounds of the underlying object.
- PointerType *AddrTy = cast<PointerType>(V->getType());
- uint64_t LoadSize = TD->getTypeStoreSize(AddrTy->getElementType());
- if (ByteOffset + LoadSize <= TD->getTypeAllocSize(BaseType) &&
- (Align == 0 || (ByteOffset % Align) == 0))
+ if (ByteOffset + LoadSize <= DL.getTypeAllocSize(BaseType) &&
+ ((ByteOffset % Align) == 0))
return true;
}
}
// from/to. If so, the previous load or store would have already trapped,
// so there is no harm doing an extra load (also, CSE will later eliminate
// the load entirely).
- BasicBlock::iterator BBI = ScanFrom, E = ScanFrom->getParent()->begin();
+ BasicBlock::iterator BBI = ScanFrom->getIterator(),
+ E = ScanFrom->getParent()->begin();
+
+ // We can at least always strip pointer casts even though we can't use the
+ // base here.
+ V = V->stripPointerCasts();
while (BBI != E) {
--BBI;
!isa<DbgInfoIntrinsic>(BBI))
return false;
+ Value *AccessedPtr;
+ unsigned AccessedAlign;
if (LoadInst *LI = dyn_cast<LoadInst>(BBI)) {
- if (AreEquivalentAddressValues(LI->getOperand(0), V)) return true;
+ AccessedPtr = LI->getPointerOperand();
+ AccessedAlign = LI->getAlignment();
} else if (StoreInst *SI = dyn_cast<StoreInst>(BBI)) {
- if (AreEquivalentAddressValues(SI->getOperand(1), V)) return true;
- }
+ AccessedPtr = SI->getPointerOperand();
+ AccessedAlign = SI->getAlignment();
+ } else
+ continue;
+
+ Type *AccessedTy = AccessedPtr->getType()->getPointerElementType();
+ if (AccessedAlign == 0)
+ AccessedAlign = DL.getABITypeAlignment(AccessedTy);
+ if (AccessedAlign < Align)
+ continue;
+
+ // Handle trivial cases.
+ if (AccessedPtr == V)
+ return true;
+
+ if (AreEquivalentAddressValues(AccessedPtr->stripPointerCasts(), V) &&
+ LoadSize <= DL.getTypeStoreSize(AccessedTy))
+ return true;
}
return false;
}
+/// DefMaxInstsToScan - the default number of maximum instructions
+/// to scan in the block, used by FindAvailableLoadedValue().
+/// FindAvailableLoadedValue() was introduced in r60148, to improve jump
+/// threading in part by eliminating partially redundant loads.
+/// At that point, the value of MaxInstsToScan was already set to '6'
+/// without documented explanation.
+cl::opt<unsigned>
+llvm::DefMaxInstsToScan("available-load-scan-limit", cl::init(6), cl::Hidden,
+ cl::desc("Use this to specify the default maximum number of instructions "
+ "to scan backward from a given instruction, when searching for "
+ "available loaded value"));
+
/// \brief Scan the ScanBB block backwards to see if we have the value at the
/// memory address *Ptr locally available within a small number of instructions.
///
if (MaxInstsToScan == 0)
MaxInstsToScan = ~0U;
- // If we're using alias analysis to disambiguate get the size of *Ptr.
- uint64_t AccessSize = 0;
- if (AA) {
- Type *AccessTy = cast<PointerType>(Ptr->getType())->getElementType();
- AccessSize = AA->getTypeStoreSize(AccessTy);
- }
+ Type *AccessTy = cast<PointerType>(Ptr->getType())->getElementType();
+
+ const DataLayout &DL = ScanBB->getModule()->getDataLayout();
+
+ // Try to get the store size for the type.
+ uint64_t AccessSize = DL.getTypeStoreSize(AccessTy);
+
+ Value *StrippedPtr = Ptr->stripPointerCasts();
while (ScanFrom != ScanBB->begin()) {
// We must ignore debug info directives when counting (otherwise they
// would affect codegen).
- Instruction *Inst = --ScanFrom;
+ Instruction *Inst = &*--ScanFrom;
if (isa<DbgInfoIntrinsic>(Inst))
continue;
// (This is true even if the load is volatile or atomic, although
// those cases are unlikely.)
if (LoadInst *LI = dyn_cast<LoadInst>(Inst))
- if (AreEquivalentAddressValues(LI->getOperand(0), Ptr)) {
+ if (AreEquivalentAddressValues(
+ LI->getPointerOperand()->stripPointerCasts(), StrippedPtr) &&
+ CastInst::isBitOrNoopPointerCastable(LI->getType(), AccessTy, DL)) {
if (AATags)
LI->getAAMetadata(*AATags);
return LI;
}
if (StoreInst *SI = dyn_cast<StoreInst>(Inst)) {
+ Value *StorePtr = SI->getPointerOperand()->stripPointerCasts();
// If this is a store through Ptr, the value is available!
// (This is true even if the store is volatile or atomic, although
// those cases are unlikely.)
- if (AreEquivalentAddressValues(SI->getOperand(1), Ptr)) {
+ if (AreEquivalentAddressValues(StorePtr, StrippedPtr) &&
+ CastInst::isBitOrNoopPointerCastable(SI->getValueOperand()->getType(),
+ AccessTy, DL)) {
if (AATags)
SI->getAAMetadata(*AATags);
return SI->getOperand(0);
}
- // If Ptr is an alloca and this is a store to a different alloca, ignore
- // the store. This is a trivial form of alias analysis that is important
- // for reg2mem'd code.
- if ((isa<AllocaInst>(Ptr) || isa<GlobalVariable>(Ptr)) &&
- (isa<AllocaInst>(SI->getOperand(1)) ||
- isa<GlobalVariable>(SI->getOperand(1))))
+ // If both StrippedPtr and StorePtr reach all the way to an alloca or
+ // global and they are different, ignore the store. This is a trivial form
+ // of alias analysis that is important for reg2mem'd code.
+ if ((isa<AllocaInst>(StrippedPtr) || isa<GlobalVariable>(StrippedPtr)) &&
+ (isa<AllocaInst>(StorePtr) || isa<GlobalVariable>(StorePtr)) &&
+ StrippedPtr != StorePtr)
continue;
// If we have alias analysis and it says the store won't modify the loaded
// value, ignore the store.
- if (AA &&
- (AA->getModRefInfo(SI, Ptr, AccessSize) & AliasAnalysis::Mod) == 0)
+ if (AA && (AA->getModRefInfo(SI, StrippedPtr, AccessSize) & MRI_Mod) == 0)
continue;
// Otherwise the store that may or may not alias the pointer, bail out.
// If alias analysis claims that it really won't modify the load,
// ignore it.
if (AA &&
- (AA->getModRefInfo(Inst, Ptr, AccessSize) & AliasAnalysis::Mod) == 0)
+ (AA->getModRefInfo(Inst, StrippedPtr, AccessSize) & MRI_Mod) == 0)
continue;
// May modify the pointer, bail out.
projects / oota-llvm.git / blobdiff