#define DEBUG_TYPE "memcpyopt"
#include "llvm/Transforms/Scalar.h"
-#include "llvm/GlobalVariable.h"
-#include "llvm/IRBuilder.h"
-#include "llvm/Instructions.h"
-#include "llvm/IntrinsicInst.h"
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/Analysis/AliasAnalysis.h"
#include "llvm/Analysis/Dominators.h"
#include "llvm/Analysis/MemoryDependenceAnalysis.h"
#include "llvm/Analysis/ValueTracking.h"
+#include "llvm/IR/DataLayout.h"
+#include "llvm/IR/GlobalVariable.h"
+#include "llvm/IR/IRBuilder.h"
+#include "llvm/IR/Instructions.h"
+#include "llvm/IR/IntrinsicInst.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/GetElementPtrTypeIterator.h"
#include "llvm/Support/raw_ostream.h"
-#include "llvm/Target/TargetData.h"
#include "llvm/Target/TargetLibraryInfo.h"
#include "llvm/Transforms/Utils/Local.h"
#include <list>
STATISTIC(NumCpyToSet, "Number of memcpys converted to memset");
static int64_t GetOffsetFromIndex(const GEPOperator *GEP, unsigned Idx,
- bool &VariableIdxFound, const TargetData &TD){
+ bool &VariableIdxFound, const DataLayout &TD){
// Skip over the first indices.
gep_type_iterator GTI = gep_type_begin(GEP);
for (unsigned i = 1; i != Idx; ++i, ++GTI)
/// constant offset, and return that constant offset. For example, Ptr1 might
/// be &A[42], and Ptr2 might be &A[40]. In this case offset would be -8.
static bool IsPointerOffset(Value *Ptr1, Value *Ptr2, int64_t &Offset,
- const TargetData &TD) {
+ const DataLayout &TD) {
Ptr1 = Ptr1->stripPointerCasts();
Ptr2 = Ptr2->stripPointerCasts();
GEPOperator *GEP1 = dyn_cast<GEPOperator>(Ptr1);
/// TheStores - The actual stores that make up this range.
SmallVector<Instruction*, 16> TheStores;
- bool isProfitableToUseMemset(const TargetData &TD) const;
+ bool isProfitableToUseMemset(const DataLayout &TD) const;
};
} // end anon namespace
-bool MemsetRange::isProfitableToUseMemset(const TargetData &TD) const {
+bool MemsetRange::isProfitableToUseMemset(const DataLayout &TD) const {
// If we found more than 4 stores to merge or 16 bytes, use memset.
if (TheStores.size() >= 4 || End-Start >= 16) return true;
// pessimize the llvm optimizer.
//
// Since we don't have perfect knowledge here, make some assumptions: assume
- // the maximum GPR width is the same size as the pointer size and assume that
- // this width can be stored. If so, check to see whether we will end up
- // actually reducing the number of stores used.
+ // the maximum GPR width is the same size as the largest legal integer
+ // size. If so, check to see whether we will end up actually reducing the
+ // number of stores used.
unsigned Bytes = unsigned(End-Start);
- unsigned NumPointerStores = Bytes/TD.getPointerSize();
+ unsigned MaxIntSize = TD.getLargestLegalIntTypeSize();
+ if (MaxIntSize == 0)
+ MaxIntSize = 1;
+ unsigned NumPointerStores = Bytes / MaxIntSize;
// Assume the remaining bytes if any are done a byte at a time.
- unsigned NumByteStores = Bytes - NumPointerStores*TD.getPointerSize();
+ unsigned NumByteStores = Bytes - NumPointerStores * MaxIntSize;
// If we will reduce the # stores (according to this heuristic), do the
// transformation. This encourages merging 4 x i8 -> i32 and 2 x i16 -> i32
/// because each element is relatively large and expensive to copy.
std::list<MemsetRange> Ranges;
typedef std::list<MemsetRange>::iterator range_iterator;
- const TargetData &TD;
+ const DataLayout &TD;
public:
- MemsetRanges(const TargetData &td) : TD(td) {}
+ MemsetRanges(const DataLayout &td) : TD(td) {}
typedef std::list<MemsetRange>::const_iterator const_iterator;
const_iterator begin() const { return Ranges.begin(); }
class MemCpyOpt : public FunctionPass {
MemoryDependenceAnalysis *MD;
TargetLibraryInfo *TLI;
- const TargetData *TD;
+ const DataLayout *TD;
public:
static char ID; // Pass identification, replacement for typeid
MemCpyOpt() : FunctionPass(ID) {
AMemSet->setDebugLoc(Range.TheStores[0]->getDebugLoc());
// Zap all the stores.
- for (SmallVector<Instruction*, 16>::const_iterator
+ for (SmallVectorImpl<Instruction *>::const_iterator
SI = Range.TheStores.begin(),
SE = Range.TheStores.end(); SI != SE; ++SI) {
MD->removeInstruction(*SI);
if (cpyLen < srcSize)
return false;
- // Check that dest points to memory that is at least as aligned as src.
- unsigned srcAlign = srcAlloca->getAlignment();
- if (!srcAlign)
- srcAlign = TD->getABITypeAlignment(srcAlloca->getAllocatedType());
- bool isDestSufficientlyAligned = srcAlign <= cpyAlign;
- // If dest is not aligned enough and we can't increase its alignment then
- // bail out.
- if (!isDestSufficientlyAligned && !isa<AllocaInst>(cpyDest))
- return false;
-
// Check that accessing the first srcSize bytes of dest will not cause a
// trap. Otherwise the transform is invalid since it might cause a trap
// to occur earlier than it otherwise would.
return false;
Type *StructTy = cast<PointerType>(A->getType())->getElementType();
- uint64_t destSize = TD->getTypeAllocSize(StructTy);
+ if (!StructTy->isSized()) {
+ // The call may never return and hence the copy-instruction may never
+ // be executed, and therefore it's not safe to say "the destination
+ // has at least <cpyLen> bytes, as implied by the copy-instruction",
+ return false;
+ }
+ uint64_t destSize = TD->getTypeAllocSize(StructTy);
if (destSize < srcSize)
return false;
} else {
return false;
}
+ // Check that dest points to memory that is at least as aligned as src.
+ unsigned srcAlign = srcAlloca->getAlignment();
+ if (!srcAlign)
+ srcAlign = TD->getABITypeAlignment(srcAlloca->getAllocatedType());
+ bool isDestSufficientlyAligned = srcAlign <= cpyAlign;
+ // If dest is not aligned enough and we can't increase its alignment then
+ // bail out.
+ if (!isDestSufficientlyAligned && !isa<AllocaInst>(cpyDest))
+ return false;
+
// Check that src is not accessed except via the call and the memcpy. This
// guarantees that it holds only undefined values when passed in (so the final
// memcpy can be dropped), that it is not read or written between the call and
bool changedArgument = false;
for (unsigned i = 0; i < CS.arg_size(); ++i)
if (CS.getArgument(i)->stripPointerCasts() == cpySrc) {
- if (cpySrc->getType() != cpyDest->getType())
- cpyDest = CastInst::CreatePointerCast(cpyDest, cpySrc->getType(),
- cpyDest->getName(), C);
+ Value *Dest = cpySrc->getType() == cpyDest->getType() ? cpyDest
+ : CastInst::CreatePointerCast(cpyDest, cpySrc->getType(),
+ cpyDest->getName(), C);
changedArgument = true;
- if (CS.getArgument(i)->getType() == cpyDest->getType())
- CS.setArgument(i, cpyDest);
+ if (CS.getArgument(i)->getType() == Dest->getType())
+ CS.setArgument(i, Dest);
else
- CS.setArgument(i, CastInst::CreatePointerCast(cpyDest,
- CS.getArgument(i)->getType(), cpyDest->getName(), C));
+ CS.setArgument(i, CastInst::CreatePointerCast(Dest,
+ CS.getArgument(i)->getType(), Dest->getName(), C));
}
if (!changedArgument)
bool MemCpyOpt::runOnFunction(Function &F) {
bool MadeChange = false;
MD = &getAnalysis<MemoryDependenceAnalysis>();
- TD = getAnalysisIfAvailable<TargetData>();
+ TD = getAnalysisIfAvailable<DataLayout>();
TLI = &getAnalysis<TargetLibraryInfo>();
// If we don't have at least memset and memcpy, there is little point of doing