[llvm] Reland "[Transforms] LoopIdiomRecognize recognize strlen and wcslen (#108985)" (PR #131412)

[Nikita Popov via llvm-commits]

================
@@ -1529,6 +1556,279 @@ static Value *matchCondition(BranchInst *BI, BasicBlock *LoopEntry,
   return nullptr;
 }
 
+namespace {
+
+class StrlenVerifier {
+public:
+  explicit StrlenVerifier(const Loop *CurLoop, ScalarEvolution *SE,
+                          const TargetLibraryInfo *TLI)
+      : CurLoop(CurLoop), SE(SE), TLI(TLI) {}
+
+  bool isValidStrlenIdiom() {
+    // Give up if the loop has multiple blocks, multiple backedges, or
+    // multiple exit blocks
+    if (CurLoop->getNumBackEdges() != 1 || CurLoop->getNumBlocks() != 1 ||
+        !CurLoop->getUniqueExitBlock())
+      return false;
+
+    // It should have a preheader and a branch instruction.
+    BasicBlock *Preheader = CurLoop->getLoopPreheader();
+    if (!Preheader)
+      return false;
+
+    BranchInst *EntryBI = dyn_cast<BranchInst>(Preheader->getTerminator());
+    if (!EntryBI)
+      return false;
+
+    // The loop exit must be conditioned on an icmp with 0 the null terminator.
+    // The icmp operand has to be a load on some SSA reg that increments
+    // by 1 in the loop.
+    BasicBlock *LoopBody = *CurLoop->block_begin();
+
+    // Skip if the body is too big as it most likely is not a strlen idiom.
+    if (!LoopBody || LoopBody->size() >= 15)
+      return false;
+
+    BranchInst *LoopTerm = dyn_cast<BranchInst>(LoopBody->getTerminator());
+    Value *LoopCond = matchCondition(LoopTerm, LoopBody);
+    if (!LoopCond)
+      return false;
+
+    LoadInst *LoopLoad = dyn_cast<LoadInst>(LoopCond);
+    if (!LoopLoad || LoopLoad->getPointerAddressSpace() != 0)
+      return false;
+
+    OperandType = LoopLoad->getType();
+    if (!OperandType || !OperandType->isIntegerTy())
+      return false;
+
+    // See if the pointer expression is an AddRec with constant step a of form
+    // ({n,+,a}) where a is the width of the char type.
+    Value *IncPtr = LoopLoad->getPointerOperand();
+    const SCEVAddRecExpr *LoadEv =
+        dyn_cast<SCEVAddRecExpr>(SE->getSCEV(IncPtr));
+    if (!LoadEv || LoadEv->getLoop() != CurLoop || !LoadEv->isAffine())
+      return false;
+    LoadBaseEv = LoadEv->getStart();
+
+    LLVM_DEBUG({
+      dbgs() << "pointer load scev: ";
+      LoadEv->print(outs());
+      dbgs() << "\n";
+    });
+
+    const SCEVConstant *Step =
+        dyn_cast<SCEVConstant>(LoadEv->getStepRecurrence(*SE));
+    if (!Step)
+      return false;
+
+    unsigned StepSize = 0;
+    StepSizeCI = dyn_cast<ConstantInt>(Step->getValue());
+    if (!StepSizeCI)
+      return false;
+    StepSize = StepSizeCI->getZExtValue();
+
+    // Verify that StepSize is consistent with platform char width.
+    OpWidth = OperandType->getIntegerBitWidth();
+    unsigned WcharSize = TLI->getWCharSize(*LoopLoad->getModule());
+    if (OpWidth != StepSize * 8)
+      return false;
+    if (OpWidth != 8 && OpWidth != 16 && OpWidth != 32)
+      return false;
+    if (OpWidth >= 16)
+      if (OpWidth != WcharSize * 8)
+        return false;
+
+    // Scan every instruction in the loop to ensure there are no side effects.
+    for (Instruction &I : *LoopBody)
+      if (I.mayHaveSideEffects())
+        return false;
+
+    BasicBlock *LoopExitBB = CurLoop->getExitBlock();
+    if (!LoopExitBB)
+      return false;
+
+    for (PHINode &PN : LoopExitBB->phis()) {
+      if (!SE->isSCEVable(PN.getType()))
+        return false;
+
+      const SCEV *Ev = SE->getSCEV(&PN);
+      if (!Ev)
+        return false;
+
+      LLVM_DEBUG({
+        dbgs() << "loop exit phi scev: ";
+        Ev->print(dbgs());
+        dbgs() << "\n";
----------------
nikic wrote:

Can use `<< *Ev`, no need for print().

https://github.com/llvm/llvm-project/pull/131412