[libc-commits] [libc] [libc] Add fuzzers for `memcpy` and `memset` (PR #90591)

[Clement Courbet via libc-commits]

================
@@ -0,0 +1,56 @@
+//===-- memcpy_fuzz.cpp ---------------------------------------------------===//
+//
+// 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
+//
+//===----------------------------------------------------------------------===//
+///
+/// Fuzzing test for llvm-libc memcpy implementation.
+///
+//===----------------------------------------------------------------------===//
+#include "protected_pages.h"
+#include "src/string/memcpy.h"
+#include <stddef.h> // size_t
+#include <stdint.h> // uint8_t
+#include <stdlib.h> // rand
+
+extern "C" int LLVMFuzzerTestOneInput(const uint8_t *data, size_t data_size) {
+  static constexpr size_t MAX_SIZE = 1024;
+  static ProtectedPages pages;
+  static const Page write_buffer = pages.GetPageA().WithAccess(PROT_WRITE);
+  static const Page read_buffer = [&]() {
+    // We fetch page B in write mode.
+    auto page = pages.GetPageB().WithAccess(PROT_WRITE);
+    // And fill it with random numbers.
+    for (size_t i = 0; i < page.page_size; ++i)
+      page.page_ptr[i] = rand();
+    // Then return it in read mode.
+    return page.WithAccess(PROT_READ);
+  }();
+  // We fill 'size' with data coming from lib_fuzzer, this limits exploration to
----------------
legrosbuffle wrote:

Same comment here: we have only 1024 different values, why not use an exhaustive loop ?

https://github.com/llvm/llvm-project/pull/90591