src, deps: add nbytes library

Projects that seek to implement Node.js compatible APIs end up needed to reproduce various bits of functionality internally in order to faithfully replicate the Node.js behaviors. This is particularly true for things like byte manipulation, base64 and hex encoding, and other low-level operations. This change proposes moving much of this low-level byte manipulation code out of nodejs/src and into a new `nbytes` library. Initially this new library will exist in the `deps` directory but the intent is to spin out a new separate repository to be its home in the future. Doing so will allow other projects to use the nbytes library with exactly the same implementation as Node.js. This commit moves only the byte swapping and legacy base64 handling code. Additional commits will move additional byte manipulation logic into the library. PR-URL: #53507 Reviewed-By: Yagiz Nizipli <yagiz.nizipli@sentry.io> Reviewed-By: Robert Nagy <ronagy@icloud.com> Reviewed-By: Chengzhong Wu <legendecas@gmail.com>
nodejs · Jun 25, 2024 · 70bbc02 · 70bbc02
1 parent e95af74
commit 70bbc02
Show file tree

Hide file tree

Showing 27 changed files with 669 additions and 593 deletions.
diff --git a/Makefile b/Makefile
@@ -174,7 +174,7 @@ with-code-cache test-code-cache:
 
 out/Makefile: config.gypi common.gypi common_node.gypi node.gyp \
 	deps/uv/uv.gyp deps/llhttp/llhttp.gyp deps/zlib/zlib.gyp \
-	deps/simdutf/simdutf.gyp deps/ada/ada.gyp \
+	deps/simdutf/simdutf.gyp deps/ada/ada.gyp deps/nbytes/nbytes.gyp \
 	tools/v8_gypfiles/toolchain.gypi tools/v8_gypfiles/features.gypi \
 	tools/v8_gypfiles/inspector.gypi tools/v8_gypfiles/v8.gyp
 	$(PYTHON) tools/gyp_node.py -f make

diff --git a/deps/nbytes/README.md b/deps/nbytes/README.md
@@ -0,0 +1,5 @@
+# Node.js bytes (nbytes) library
+
+The `nbytes` library extracts certain Node.js specific byte manipulation
+functions from the core of Node.js itself and makes them available for
+use in other projects that need to emulate Node.js' behavior.
diff --git a/deps/nbytes/nbytes.cpp b/deps/nbytes/nbytes.cpp
@@ -0,0 +1,251 @@
+#include "nbytes.h"
+#include <string.h>
+#include <cmath>
+#include <cstddef>
+
+namespace nbytes {
+
+// ============================================================================
+// Byte Swapping
+
+namespace {
+// These are defined by <sys/byteorder.h> or <netinet/in.h> on some systems.
+// To avoid warnings, undefine them before redefining them.
+#ifdef BSWAP_2
+# undef BSWAP_2
+#endif
+#ifdef BSWAP_4
+# undef BSWAP_4
+#endif
+#ifdef BSWAP_8
+# undef BSWAP_8
+#endif
+
+#if defined(_MSC_VER)
+#include <intrin.h>
+#define BSWAP_2(x) _byteswap_ushort(x)
+#define BSWAP_4(x) _byteswap_ulong(x)
+#define BSWAP_8(x) _byteswap_uint64(x)
+#else
+#define BSWAP_2(x) ((x) << 8) | ((x) >> 8)
+#define BSWAP_4(x)                                                            \
+  (((x) & 0xFF) << 24) |                                                      \
+  (((x) & 0xFF00) << 8) |                                                     \
+  (((x) >> 8) & 0xFF00) |                                                     \
+  (((x) >> 24) & 0xFF)
+#define BSWAP_8(x)                                                            \
+  (((x) & 0xFF00000000000000ull) >> 56) |                                     \
+  (((x) & 0x00FF000000000000ull) >> 40) |                                     \
+  (((x) & 0x0000FF0000000000ull) >> 24) |                                     \
+  (((x) & 0x000000FF00000000ull) >> 8) |                                      \
+  (((x) & 0x00000000FF000000ull) << 8) |                                      \
+  (((x) & 0x0000000000FF0000ull) << 24) |                                     \
+  (((x) & 0x000000000000FF00ull) << 40) |                                     \
+  (((x) & 0x00000000000000FFull) << 56)
+#endif
+}  // namespace
+
+bool SwapBytes16(void* data, size_t nbytes) {
+  if (nbytes % sizeof(uint16_t) != 0) return false;
+
+#if defined(_MSC_VER)
+  if (AlignUp(data, sizeof(uint16_t)) == data) {
+    // MSVC has no strict aliasing, and is able to highly optimize this case.
+    uint16_t* data16 = reinterpret_cast<uint16_t*>(data);
+    size_t len16 = nbytes / sizeof(uint16_t);
+    for (size_t i = 0; i < len16; i++) {
+      data16[i] = BSWAP_2(data16[i]);
+    }
+    return true;
+  }
+#endif
+
+  uint16_t temp;
+  uint8_t* ptr = reinterpret_cast<uint8_t*>(data);
+  for (size_t i = 0; i < nbytes; i += sizeof(uint16_t)) {
+    memcpy(&temp, &ptr[i], sizeof(uint16_t));
+    temp = BSWAP_2(temp);
+    memcpy(&ptr[i], &temp, sizeof(uint16_t));
+  }
+
+  return true;
+}
+
+bool SwapBytes32(void* data, size_t nbytes) {
+  if (nbytes % sizeof(uint32_t) != 0) return false;
+
+#if defined(_MSC_VER)
+  // MSVC has no strict aliasing, and is able to highly optimize this case.
+  if (AlignUp(data, sizeof(uint32_t)) == data) {
+    uint32_t* data32 = reinterpret_cast<uint32_t*>(data);
+    size_t len32 = nbytes / sizeof(uint32_t);
+    for (size_t i = 0; i < len32; i++) {
+      data32[i] = BSWAP_4(data32[i]);
+    }
+    return true;
+  }
+#endif
+
+  uint32_t temp = 0;
+  uint8_t* ptr = reinterpret_cast<uint8_t*>(data);
+  for (size_t i = 0; i < nbytes; i += sizeof(uint32_t)) {
+    memcpy(&temp, &ptr[i], sizeof(uint32_t));
+    temp = BSWAP_4(temp);
+    memcpy(&ptr[i], &temp, sizeof(uint32_t));
+  }
+
+  return true;
+}
+
+bool SwapBytes64(void* data, size_t nbytes) {
+  if (nbytes % sizeof(uint64_t) != 0) return false;
+
+#if defined(_MSC_VER)
+  if (AlignUp(data, sizeof(uint64_t)) == data) {
+    // MSVC has no strict aliasing, and is able to highly optimize this case.
+    uint64_t* data64 = reinterpret_cast<uint64_t*>(data);
+    size_t len64 = nbytes / sizeof(uint64_t);
+    for (size_t i = 0; i < len64; i++) {
+      data64[i] = BSWAP_8(data64[i]);
+    }
+    return true;
+  }
+#endif
+
+  uint64_t temp = 0;
+  uint8_t* ptr = reinterpret_cast<uint8_t*>(data);
+  for (size_t i = 0; i < nbytes; i += sizeof(uint64_t)) {
+    memcpy(&temp, &ptr[i], sizeof(uint64_t));
+    temp = BSWAP_8(temp);
+    memcpy(&ptr[i], &temp, sizeof(uint64_t));
+  }
+
+  return true;
+}
+
+// ============================================================================
+// Base64 (legacy)
+
+// supports regular and URL-safe base64
+const int8_t unbase64_table[256] =
+  { -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -2, -1, -1, -2, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -2, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, 62, -1, 62, -1, 63,
+    52, 53, 54, 55, 56, 57, 58, 59, 60, 61, -1, -1, -1, -1, -1, -1,
+    -1,  0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14,
+    15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, -1, -1, -1, -1, 63,
+    -1, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40,
+    41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1
+  };
+
+// ============================================================================
+// Hex
+
+const int8_t unhex_table[256] =
+  { -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+     0,  1,  2,  3,  4,  5,  6,  7,  8,  9, -1, -1, -1, -1, -1, -1,
+    -1, 10, 11, 12, 13, 14, 15, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, 10, 11, 12, 13, 14, 15, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1
+  };
+
+size_t HexEncode(
+    const char* src,
+    size_t slen,
+    char* dst,
+    size_t dlen) {
+  // We know how much we'll write, just make sure that there's space.
+  NBYTES_ASSERT_TRUE(
+      dlen >= MultiplyWithOverflowCheck<size_t>(slen, 2u) &&
+      "not enough space provided for hex encode");
+
+  dlen = slen * 2;
+  for (size_t i = 0, k = 0; k < dlen; i += 1, k += 2) {
+    static const char hex[] = "0123456789abcdef";
+    uint8_t val = static_cast<uint8_t>(src[i]);
+    dst[k + 0] = hex[val >> 4];
+    dst[k + 1] = hex[val & 15];
+  }
+
+  return dlen;
+}
+
+std::string HexEncode(const char* src, size_t slen) {
+  size_t dlen = slen * 2;
+  std::string dst(dlen, '\0');
+  HexEncode(src, slen, dst.data(), dlen);
+  return dst;
+}
+
+// ============================================================================
+
+void ForceAsciiSlow(const char* src, char* dst, size_t len) {
+  for (size_t i = 0; i < len; ++i) {
+    dst[i] = src[i] & 0x7f;
+  }
+}
+
+void ForceAscii(const char* src, char* dst, size_t len) {
+  if (len < 16) {
+    ForceAsciiSlow(src, dst, len);
+    return;
+  }
+
+  const unsigned bytes_per_word = sizeof(uintptr_t);
+  const unsigned align_mask = bytes_per_word - 1;
+  const unsigned src_unalign = reinterpret_cast<uintptr_t>(src) & align_mask;
+  const unsigned dst_unalign = reinterpret_cast<uintptr_t>(dst) & align_mask;
+
+  if (src_unalign > 0) {
+    if (src_unalign == dst_unalign) {
+      const unsigned unalign = bytes_per_word - src_unalign;
+      ForceAsciiSlow(src, dst, unalign);
+      src += unalign;
+      dst += unalign;
+      len -= src_unalign;
+    } else {
+      ForceAsciiSlow(src, dst, len);
+      return;
+    }
+  }
+
+#if defined(_WIN64) || defined(_LP64)
+  const uintptr_t mask = ~0x8080808080808080ll;
+#else
+  const uintptr_t mask = ~0x80808080l;
+#endif
+
+  const uintptr_t* srcw = reinterpret_cast<const uintptr_t*>(src);
+  uintptr_t* dstw = reinterpret_cast<uintptr_t*>(dst);
+
+  for (size_t i = 0, n = len / bytes_per_word; i < n; ++i) {
+    dstw[i] = srcw[i] & mask;
+  }
+
+  const unsigned remainder = len & align_mask;
+  if (remainder > 0) {
+    const size_t offset = len - remainder;
+    ForceAsciiSlow(src + offset, dst + offset, remainder);
+  }
+}
+
+}  // namespace nbytes
diff --git a/deps/nbytes/nbytes.gyp b/deps/nbytes/nbytes.gyp
@@ -0,0 +1,16 @@
+{
+  'variables': {
+    'nbytes_sources': [ 'nbytes.cpp' ],
+  },
+  'targets': [
+    {
+      'target_name': 'nbytes',
+      'type': 'static_library',
+      'include_dirs': ['.'],
+      'direct_dependent_settings': {
+        'include_dirs': ['.'],
+      },
+      'sources': [ '<@(nbytes_sources)' ]
+    },
+  ]
+}