[llvm] Use DIExpression::foldConstantMath() at the result of an append() (PR #71719)

[Adrian Prantl via llvm-commits]

================
@@ -0,0 +1,349 @@
+//===- DIExpressionOptimizer.cpp - Constant folding of DIExpressions ------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements functions to constant fold DIExpressions. Which were
+// declared in DIExpressionOptimizer.h
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/BinaryFormat/Dwarf.h"
+#include "llvm/IR/DebugInfoMetadata.h"
+
+using namespace llvm;
+
+static std::optional<uint64_t> isConstantVal(DIExpression::ExprOperand Op) {
+  if (Op.getOp() == dwarf::DW_OP_constu)
+    return Op.getArg(0);
+  return std::nullopt;
+}
+
+static bool isNeutralElement(uint64_t Op, uint64_t Val) {
+  switch (Op) {
+  case dwarf::DW_OP_plus:
+  case dwarf::DW_OP_minus:
+  case dwarf::DW_OP_shl:
+  case dwarf::DW_OP_shr:
+    return Val == 0;
+  case dwarf::DW_OP_mul:
+  case dwarf::DW_OP_div:
+    return Val == 1;
+  default:
+    return false;
+  }
+}
+
+static std::optional<uint64_t> foldOperationIfPossible(
----------------
adrian-prantl wrote:

This function needs a few comments explaining at a high level what's happening. Especially the inout bool looks suspicious from a design perspective.

https://github.com/llvm/llvm-project/pull/71719