Verify integrity of memory initializer using CRC-32

This is almost the standard CRC-32 algorithm, except that we omit the final
XOR with -1 so that we can easily compare the result against zero.
The length of the initializer is included in the data so that we don't have
to worry about leading zeros (after XOR with the init value of -1).

Useful read: http://www.ross.net/crc/download/crc_v3.txt

emcc

@@ -1357,14 +1357,28 @@ try:
           membytes.pop()
         if not membytes:
           return '';
-      membytes = ''.join(map(chr, membytes))
       if not memory_init_file:
-        s = membytes
+        crcTable = []
+        for i in range(256):
+          crc = i
+          for bit in range(8):
+            crc = (crc >> 1) ^ ((crc & 1) * 0xedb88320)
+          crcTable.append(crc)
+        crc = 0xffffffff
+        s = list(membytes)
+        n = len(s)
+        crc = crcTable[(crc ^ n) & 0xff] ^ (crc >> 8)
+        crc = crcTable[(crc ^ (n >> 8)) & 0xff] ^ (crc >> 8)
+        for i in s:
+          crc = crcTable[(crc ^ i) & 0xff] ^ (crc >> 8)
+        for i in range(4):
+          s.append((crc >> (8 * i)) & 0xff)
+        s = ''.join(map(chr, s))
         s = s.replace('\\', '\\\\').replace("'", "\\'")
         s = s.replace('\n', '\\n').replace('\r', '\\r')
         s = s.decode('latin1').encode('utf8')
         return "var memoryInitializer = '%s';" % s
-      open(memfile, 'wb').write(membytes)
+      open(memfile, 'wb').write(''.join(map(chr, membytes)))
       if DEBUG:
         # Copy into temp dir as well, so can be run there too
         shared.safe_copy(memfile, os.path.join(shared.get_emscripten_temp_dir(), os.path.basename(memfile)))

src/postamble.js

@@ -7,7 +7,23 @@ if (memoryInitializer && !ENVIRONMENT_IS_PTHREAD) (function(s) {
 #else
 if (memoryInitializer) (function(s) {
 #endif
-  for (var i = 0; i < s.length; ++i) HEAPU8[STATIC_BASE + i] = s.charCodeAt(i);
+  var i, n = s.length - 4;
+#if ASSERTIONS
+  var crc, bit, table = new Int32Array(256);
+  for (i = 0; i < 256; ++i) {
+    for (crc = i, bit = 0; bit < 8; ++bit)
+      crc = (crc >>> 1) ^ ((crc & 1) * 0xedb88320);
+    table[i] = crc >>> 0;
+  }
+  crc = -1;
+  crc = table[(crc ^ n) & 0xff] ^ (crc >>> 8);
+  crc = table[(crc ^ (n >>> 8)) & 0xff] ^ (crc >>> 8);
+  for (i = 0; i < s.length; ++i)
+    crc = table[(crc ^ s.charCodeAt(i)) & 0xff] ^ (crc >>> 8);
+  assert(crc === 0, "memory initializer checksum");
+#endif
+  for (i = 0; i < n; ++i)
+    HEAPU8[STATIC_BASE + i] = s.charCodeAt(i);
 })(memoryInitializer);
 #else
 #if MEM_INIT_METHOD == 1

tests/test_other.py

@@ -4879,3 +4879,17 @@ int main() {
     out, err = Popen([PYTHON, EMCC, 'src.c', '-s', 'EXPORTED_FUNCTIONS=["_main", "_treecount"]', '--minify', '0', '-g4', '-Oz']).communicate()
     self.assertContained('hello, world!', run_js('a.out.js'))
 
+  def test_meminit_crc(self):
+    with open('src.c', 'w') as f:
+      f.write(r'''
+#include<stdio.h>
+int main() { printf("Mary had a little lamb.\n"); }
+''')
+    out, err = Popen([PYTHON, EMCC, 'src.c', '-O2', '--memory-init-file', '0', '-s', 'MEM_INIT_METHOD=2', '-s', 'ASSERTIONS=1']).communicate()
+    with open('a.out.js', 'r') as f:
+      d = f.read()
+    d = d.replace('Mary had', 'Paul had')
+    with open('a.out.js', 'w') as f:
+      f.write(d)
+    out = run_js('a.out.js', assert_returncode=None, stderr=subprocess.STDOUT)
+    self.assertContained('Assertion failed: memory initializer checksum', out)