bool processMemMove(MemMoveInst *M);
bool performCallSlotOptzn(Instruction *cpy, Value *cpyDst, Value *cpySrc,
uint64_t cpyLen, unsigned cpyAlign, CallInst *C);
- bool processMemCpyMemCpyDependence(MemCpyInst *M, MemCpyInst *MDep,
- uint64_t MSize);
+ bool processMemCpyMemCpyDependence(MemCpyInst *M, MemCpyInst *MDep);
bool processMemSetMemCpyDependence(MemCpyInst *M, MemSetInst *MDep);
+ bool performMemCpyToMemSetOptzn(MemCpyInst *M, MemSetInst *MDep);
bool processByValArgument(CallSite CS, unsigned ArgNo);
Instruction *tryMergingIntoMemset(Instruction *I, Value *StartPtr,
Value *ByteVal);
// Check that nothing touches the dest of the "copy" between
// the call and the store.
AliasAnalysis &AA = getAnalysis<AliasAnalysis>();
- AliasAnalysis::Location StoreLoc = AA.getLocation(SI);
+ AliasAnalysis::Location StoreLoc = MemoryLocation::get(SI);
for (BasicBlock::iterator I = --BasicBlock::iterator(SI),
E = C; I != E; --I) {
if (AA.getModRefInfo(&*I, StoreLoc) != AliasAnalysis::NoModRef) {
/// processMemCpyMemCpyDependence - We've found that the (upward scanning)
/// memory dependence of memcpy 'M' is the memcpy 'MDep'. Try to simplify M to
-/// copy from MDep's input if we can. MSize is the size of M's copy.
+/// copy from MDep's input if we can.
///
-bool MemCpyOpt::processMemCpyMemCpyDependence(MemCpyInst *M, MemCpyInst *MDep,
- uint64_t MSize) {
+bool MemCpyOpt::processMemCpyMemCpyDependence(MemCpyInst *M, MemCpyInst *MDep) {
// We can only transforms memcpy's where the dest of one is the source of the
// other.
if (M->getSource() != MDep->getDest() || MDep->isVolatile())
//
// NOTE: This is conservative, it will stop on any read from the source loc,
// not just the defining memcpy.
- MemDepResult SourceDep =
- MD->getPointerDependencyFrom(AA.getLocationForSource(MDep),
- false, M, M->getParent());
+ MemDepResult SourceDep = MD->getPointerDependencyFrom(
+ MemoryLocation::getForSource(MDep), false, M, M->getParent());
if (!SourceDep.isClobber() || SourceDep.getInst() != MDep)
return false;
// source and dest might overlap. We still want to eliminate the intermediate
// value, but we have to generate a memmove instead of memcpy.
bool UseMemMove = false;
- if (!AA.isNoAlias(AA.getLocationForDest(M), AA.getLocationForSource(MDep)))
+ if (!AA.isNoAlias(MemoryLocation::getForDest(M),
+ MemoryLocation::getForSource(MDep)))
UseMemMove = true;
// If all checks passed, then we can transform M.
if (MemSet->getDest() != MemCpy->getDest())
return false;
- Value *Dest = MemSet->getDest();
+ // Check that there are no other dependencies on the memset destination.
+ MemDepResult DstDepInfo = MD->getPointerDependencyFrom(
+ MemoryLocation::getForDest(MemSet), false, MemCpy, MemCpy->getParent());
+ if (DstDepInfo.getInst() != MemSet)
+ return false;
+
+ // Use the same i8* dest as the memcpy, killing the memset dest if different.
+ Value *Dest = MemCpy->getRawDest();
Value *DestSize = MemSet->getLength();
Value *SrcSize = MemCpy->getLength();
if (ConstantInt *SrcSizeC = dyn_cast<ConstantInt>(SrcSize))
Align = MinAlign(SrcSizeC->getZExtValue(), DestAlign);
- IRBuilder<> Builder(MemCpy->getNextNode());
+ IRBuilder<> Builder(MemCpy);
// If the sizes have different types, zext the smaller one.
if (DestSize->getType() != SrcSize->getType()) {
return true;
}
+/// Transform memcpy to memset when its source was just memset.
+/// In other words, turn:
+/// \code
+/// memset(dst1, c, dst1_size);
+/// memcpy(dst2, dst1, dst2_size);
+/// \endcode
+/// into:
+/// \code
+/// memset(dst1, c, dst1_size);
+/// memset(dst2, c, dst2_size);
+/// \endcode
+/// When dst2_size <= dst1_size.
+///
+/// The \p MemCpy must have a Constant length.
+bool MemCpyOpt::performMemCpyToMemSetOptzn(MemCpyInst *MemCpy,
+ MemSetInst *MemSet) {
+ // This only makes sense on memcpy(..., memset(...), ...).
+ if (MemSet->getRawDest() != MemCpy->getRawSource())
+ return false;
+
+ ConstantInt *CopySize = cast<ConstantInt>(MemCpy->getLength());
+ ConstantInt *MemSetSize = dyn_cast<ConstantInt>(MemSet->getLength());
+ // Make sure the memcpy doesn't read any more than what the memset wrote.
+ // Don't worry about sizes larger than i64.
+ if (!MemSetSize || CopySize->getZExtValue() > MemSetSize->getZExtValue())
+ return false;
+
+ IRBuilder<> Builder(MemCpy);
+ Builder.CreateMemSet(MemCpy->getRawDest(), MemSet->getOperand(1),
+ CopySize, MemCpy->getAlignment());
+ return true;
+}
+
/// processMemCpy - perform simplification of memcpy's. If we have memcpy A
/// which copies X to Y, and memcpy B which copies Y to Z, then we can rewrite
/// B to be a memcpy from X to Z (or potentially a memmove, depending on
return true;
}
- AliasAnalysis::Location SrcLoc = AliasAnalysis::getLocationForSource(M);
- MemDepResult SrcDepInfo = MD->getPointerDependencyFrom(SrcLoc, true,
- M, M->getParent());
+ MemDepResult DepInfo = MD->getDependency(M);
// Try to turn a partially redundant memset + memcpy into
// memcpy + smaller memset. We don't need the memcpy size for this.
- if (SrcDepInfo.isClobber())
- if (MemSetInst *MDep = dyn_cast<MemSetInst>(SrcDepInfo.getInst()))
+ if (DepInfo.isClobber())
+ if (MemSetInst *MDep = dyn_cast<MemSetInst>(DepInfo.getInst()))
if (processMemSetMemCpyDependence(M, MDep))
return true;
ConstantInt *CopySize = dyn_cast<ConstantInt>(M->getLength());
if (!CopySize) return false;
- // The are three possible optimizations we can do for memcpy:
+ // There are four possible optimizations we can do for memcpy:
// a) memcpy-memcpy xform which exposes redundance for DSE.
// b) call-memcpy xform for return slot optimization.
// c) memcpy from freshly alloca'd space or space that has just started its
// lifetime copies undefined data, and we can therefore eliminate the
// memcpy in favor of the data that was already at the destination.
- MemDepResult DepInfo = MD->getDependency(M);
+ // d) memcpy from a just-memset'd source can be turned into memset.
if (DepInfo.isClobber()) {
if (CallInst *C = dyn_cast<CallInst>(DepInfo.getInst())) {
if (performCallSlotOptzn(M, M->getDest(), M->getSource(),
}
}
+ AliasAnalysis::Location SrcLoc = MemoryLocation::getForSource(M);
+ MemDepResult SrcDepInfo = MD->getPointerDependencyFrom(SrcLoc, true,
+ M, M->getParent());
+
if (SrcDepInfo.isClobber()) {
if (MemCpyInst *MDep = dyn_cast<MemCpyInst>(SrcDepInfo.getInst()))
- return processMemCpyMemCpyDependence(M, MDep, CopySize->getZExtValue());
+ return processMemCpyMemCpyDependence(M, MDep);
} else if (SrcDepInfo.isDef()) {
Instruction *I = SrcDepInfo.getInst();
bool hasUndefContents = false;
}
}
+ if (SrcDepInfo.isClobber())
+ if (MemSetInst *MDep = dyn_cast<MemSetInst>(SrcDepInfo.getInst()))
+ if (performMemCpyToMemSetOptzn(M, MDep)) {
+ MD->removeInstruction(M);
+ M->eraseFromParent();
+ ++NumCpyToSet;
+ return true;
+ }
+
return false;
}
return false;
// See if the pointers alias.
- if (!AA.isNoAlias(AA.getLocationForDest(M), AA.getLocationForSource(M)))
+ if (!AA.isNoAlias(MemoryLocation::getForDest(M),
+ MemoryLocation::getForSource(M)))
return false;
DEBUG(dbgs() << "MemCpyOpt: Optimizing memmove -> memcpy: " << *M << "\n");
// NOTE: This is conservative, it will stop on any read from the source loc,
// not just the defining memcpy.
MemDepResult SourceDep =
- MD->getPointerDependencyFrom(AliasAnalysis::getLocationForSource(MDep),
- false, CS.getInstruction(), MDep->getParent());
+ MD->getPointerDependencyFrom(MemoryLocation::getForSource(MDep), false,
+ CS.getInstruction(), MDep->getParent());
if (!SourceDep.isClobber() || SourceDep.getInst() != MDep)
return false;