[wasm] Add UTF-8 validation

src/wasm/utf8.cc

Index: src/wasm/utf8.cc
diff --git a/src/wasm/utf8.cc b/src/wasm/utf8.cc
new file mode 100644
index 0000000000000000000000000000000000000000..1ae58d4e17100639a1f7cc54f183336714d735bc
--- /dev/null
+++ b/src/wasm/utf8.cc
@@ -0,0 +1,42 @@
+// Copyright 2016 the V8 project authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file.
+
+#include "src/wasm/utf8.h"
+
+#include "src/base/logging.h"
+
+#include "unicode/utf8.h"
+
+namespace {
+
+inline bool IsValidCharacter(uint32_t code_point) {
+  // Excludes non-characters (U+FDD0..U+FDEF, and all codepoints ending in
+  // 0xFFFE or 0xFFFF) from the set of valid code points.
+  return code_point < 0xD800u ||
+         (code_point >= 0xE000u && code_point < 0xFDD0u) ||
+         (code_point > 0xFDEFu && code_point <= 0x10FFFFu &&
+          (code_point & 0xFFFEu) != 0xFFFEu);
+}
+
+}  // namespace
+
+namespace v8 {
+namespace internal {
+namespace wasm {
+
+bool IsValidUtf8(const uint8_t *buf, int32_t len) {
+  int32_t char_index = 0;
+  DCHECK(len >= 0);
+
+  while (char_index < len) {
+    int32_t code_point;
+    U8_NEXT(buf, char_index, len, code_point);
+    if (!IsValidCharacter(code_point)) return false;
+  }
+  return true;
+}
+
+}  // namespace wasm
+}  // namespace internal
+}  // namespace v8