maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

[memcpyopt] Common code into performCallSlotOptzn [NFC] 

We have the same code repeated in both callers, sink it into callee.

The motivation here isn't just code style, we can also defer the relatively expensive aliasing checks until the cheap structural preconditions have been validated.  (e.g. Don't bother aliasing if src is not an alloca.)  This helps compile time significantly.
This commit is contained in:
Philip Reames 2022-03-28 20:06:22 -07:00
parent 7d6e8f2a96
commit 33deaa13b8
1 changed files with 32 additions and 31 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

63
llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp
View File

@ -770,15 +770,6 @@ bool MemCpyOptPass::processStore(StoreInst *SI, BasicBlock::iterator &BBI) {
C = dyn_cast_or_null<CallInst>(LoadClobber->getMemoryInst());
}
if (C) {
// Check that nothing touches the dest of the "copy" between
// the call and the store.
MemoryLocation StoreLoc = MemoryLocation::get(SI);
if (accessedBetween(*AA, StoreLoc, MSSA->getMemoryAccess(C),
MSSA->getMemoryAccess(SI)))
C = nullptr;
}
if (C) {
bool changed = performCallSlotOptzn(
LI, SI, SI->getPointerOperand()->stripPointerCasts(),
@ -905,6 +896,23 @@ bool MemCpyOptPass::performCallSlotOptzn(Instruction *cpyLoad,
if (cpySize < srcSize)
return false;
if (C->getParent() != cpyStore->getParent()) {
LLVM_DEBUG(dbgs() << "Call Slot: block local restriction\n");
return false;
}
MemoryLocation DestLoc = isa<StoreInst>(cpyStore) ?
MemoryLocation::get(cpyStore) :
MemoryLocation::getForDest(cast<MemCpyInst>(cpyStore));
// Check that nothing touches the dest of the copy between
// the call and the store/memcpy.
if (accessedBetween(*AA, DestLoc, MSSA->getMemoryAccess(C),
MSSA->getMemoryAccess(cpyStore))) {
LLVM_DEBUG(dbgs() << "Call Slot: Dest pointer modified after call\n");
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.
@ -914,6 +922,7 @@ bool MemCpyOptPass::performCallSlotOptzn(Instruction *cpyLoad,
return false;
}
// Make sure that nothing can observe cpyDest being written early. There are
// a number of cases to consider:
// 1. cpyDest cannot be accessed between C and cpyStore as a precondition of
@ -1443,28 +1452,20 @@ bool MemCpyOptPass::processMemCpy(MemCpyInst *M, BasicBlock::iterator &BBI) {
if (Instruction *MI = MD->getMemoryInst()) {
if (auto *CopySize = dyn_cast<ConstantInt>(M->getLength())) {
if (auto *C = dyn_cast<CallInst>(MI)) {
// The memcpy must post-dom the call. Limit to the same block for
// now. Additionally, we need to ensure that there are no accesses
// to dest between the call and the memcpy. Accesses to src will be
// checked by performCallSlotOptzn().
// TODO: Support non-local call-slot optimization?
if (C->getParent() == M->getParent() &&
!accessedBetween(*AA, DestLoc, MD, MA)) {
// FIXME: Can we pass in either of dest/src alignment here instead
// of conservatively taking the minimum?
Align Alignment = std::min(M->getDestAlign().valueOrOne(),
M->getSourceAlign().valueOrOne());
if (performCallSlotOptzn(
M, M, M->getDest(), M->getSource(),
TypeSize::getFixed(CopySize->getZExtValue()), Alignment,
C)) {
LLVM_DEBUG(dbgs() << "Performed call slot optimization:\n"
<< " call: " << *C << "\n"
<< " memcpy: " << *M << "\n");
eraseInstruction(M);
++NumMemCpyInstr;
return true;
}
// FIXME: Can we pass in either of dest/src alignment here instead
// of conservatively taking the minimum?
Align Alignment = std::min(M->getDestAlign().valueOrOne(),
M->getSourceAlign().valueOrOne());
if (performCallSlotOptzn(
M, M, M->getDest(), M->getSource(),
TypeSize::getFixed(CopySize->getZExtValue()), Alignment,
C)) {
LLVM_DEBUG(dbgs() << "Performed call slot optimization:\n"
<< " call: " << *C << "\n"
<< " memcpy: " << *M << "\n");
eraseInstruction(M);
++NumMemCpyInstr;
return true;
}
}
}