[llvm] [AArch64][ARM] Optimize more `tbl`/`tbx` calls into `shufflevector` (PR #169748)

[David Green via llvm-commits]

================
@@ -737,42 +737,122 @@ static Instruction *foldCtpop(IntrinsicInst &II, InstCombinerImpl &IC) {
   return nullptr;
 }
 
-/// Convert a table lookup to shufflevector if the mask is constant.
-/// This could benefit tbl1 if the mask is { 7,6,5,4,3,2,1,0 }, in
-/// which case we could lower the shufflevector with rev64 instructions
-/// as it's actually a byte reverse.
-static Value *simplifyNeonTbl1(const IntrinsicInst &II,
-                               InstCombiner::BuilderTy &Builder) {
+/// Convert `tbl`/`tbx` intrinsics to shufflevector if the mask is constant, and
+/// at most two source operands are actually referenced.
+static Instruction *simplifyNeonTbl(IntrinsicInst &II, InstCombiner &IC,
+                                    bool IsExtension) {
   // Bail out if the mask is not a constant.
-  auto *C = dyn_cast<Constant>(II.getArgOperand(1));
+  auto *C = dyn_cast<Constant>(II.getArgOperand(II.arg_size() - 1));
   if (!C)
     return nullptr;
 
-  auto *VecTy = cast<FixedVectorType>(II.getType());
-  unsigned NumElts = VecTy->getNumElements();
+  auto *RetTy = cast<FixedVectorType>(II.getType());
+  unsigned NumIndexes = RetTy->getNumElements();
 
-  // Only perform this transformation for <8 x i8> vector types.
-  if (!VecTy->getElementType()->isIntegerTy(8) || NumElts != 8)
+  // Only perform this transformation for <8 x i8> and <16 x i8> vector types.
+  if (!(RetTy->getElementType()->isIntegerTy(8) &&
+        (NumIndexes == 8 || NumIndexes == 16)))
----------------
davemgreen wrote:

This can be Demorganed

https://github.com/llvm/llvm-project/pull/169748