[llvm] [InstCombine] Canonicalize complex boolean expressions into ~((y | z) ^ x) via 3-input truth table (PR #149530)

[via llvm-commits]

================
@@ -47,6 +49,202 @@ static Value *getFCmpValue(unsigned Code, Value *LHS, Value *RHS,
   return Builder.CreateFCmpFMF(NewPred, LHS, RHS, FMF);
 }
 
+/// This is to create optimal 3-variable boolean logic from truth tables.
+/// currently it supports the cases pertaining to the issue 97044. More cases
+/// can be added based on real-world justification for specific 3 input cases
+///  or with reviewer approval all 256 cases can be added (choose the
+///  canonicalizations found
+/// in x86InstCombine.cpp?)
+static Value *createLogicFromTable3Var(const std::bitset<8> &Table, Value *Op0,
+                                       Value *Op1, Value *Op2, Value *Root,
+                                       IRBuilderBase &Builder, bool HasOneUse) {
+  uint8_t TruthValue = Table.to_ulong();
+
+  // Skip transformation if expression is already simple (at most 2 levels
+  // deep).
+  if (Root->hasOneUse() && isa<BinaryOperator>(Root)) {
+    if (auto *BO = dyn_cast<BinaryOperator>(Root)) {
+      bool IsSimple = !isa<BinaryOperator>(BO->getOperand(0)) ||
+                      !isa<BinaryOperator>(BO->getOperand(1));
+      if (IsSimple)
+        return nullptr;
+    }
+  }
+
+  auto FoldConstant = [&](bool Val) {
+    Constant *Res = Val ? Builder.getTrue() : Builder.getFalse();
+    if (Op0->getType()->isVectorTy())
+      Res = ConstantVector::getSplat(
+          cast<VectorType>(Op0->getType())->getElementCount(), Res);
+    return Res;
+  };
+
+  Value *Result = nullptr;
+  switch (TruthValue) {
+  default:
+    return nullptr;
+
+  case 0x00: // Always FALSE
+    Result = FoldConstant(false);
+    break;
+
+  case 0xFF: // Always TRUE
+    Result = FoldConstant(true);
+    break;
+
+  case 0xE1: // ~((Op1 | Op2) ^ Op0)
+    if (!HasOneUse)
+      return nullptr;
+    {
+      Value *Or = Builder.CreateOr(Op1, Op2);
+      Value *Xor = Builder.CreateXor(Or, Op0);
+      Result = Builder.CreateNot(Xor);
+    }
+    break;
+
+  case 0x60: // Op0 & (Op1 ^ Op2)
+    if (!HasOneUse)
+      return nullptr;
+    {
+      Value *Xor = Builder.CreateXor(Op1, Op2);
+      Result = Builder.CreateAnd(Op0, Xor);
+    }
+    break;
+
+  case 0xD2: // ((Op1 | Op2) ^ Op0) ^ Op1
+    if (!HasOneUse)
+      return nullptr;
+    {
+      Value *Or = Builder.CreateOr(Op1, Op2);
+      Value *Xor1 = Builder.CreateXor(Or, Op0);
+      Result = Builder.CreateXor(Xor1, Op1);
+    }
+    break;
+  }
+
+  return Result;
+}
+
+static std::tuple<Value *, Value *, Value *>
+extractThreeVariables(Value *Root) {
+  std::set<Value *> Variables;
+  unsigned NodeCount = 0;
+  const unsigned MaxNodes =
+      50; // To prevent exponential blowup (see bitwise-hang.ll)
+
+  std::function<void(Value *)> Collect = [&](Value *V) {
----------------
yafet-a wrote:

Thanks for the suggestion, this has been implemented now

https://github.com/llvm/llvm-project/pull/149530