Document that its the job of the embedder to provide strong entropy for seeding the PRNG.

BUG=v8:2905
R=dslomov@chromium.org

Review URL: https://codereview.chromium.org/23965009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16888 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

src/utils/random-number-generator.cc

@@ -82,6 +82,11 @@ RandomNumberGenerator::RandomNumberGenerator() {
   // We cannot assume that random() or rand() were seeded
   // properly, so instead of relying on random() or rand(),
   // we just seed our PRNG using timing data as fallback.
+  // This is weak entropy, but it's sufficient, because
+  // it is the responsibility of the embedder to install
+  // an entropy source using v8::V8::SetEntropySource(),
+  // which provides reasonable entropy, see:
+  // https://code.google.com/p/v8/issues/detail?id=2905
   int64_t seed = Time::NowFromSystemTime().ToInternalValue() << 24;
   seed ^= TimeTicks::HighResNow().ToInternalValue() << 16;
   seed ^= TimeTicks::Now().ToInternalValue() << 8;

src/utils/random-number-generator.h

@@ -42,6 +42,10 @@ namespace internal {
 // If two instances of RandomNumberGenerator are created with the same seed, and
 // the same sequence of method calls is made for each, they will generate and
 // return identical sequences of numbers.
+// This class uses (probably) weak entropy by default, but it's sufficient,
+// because it is the responsibility of the embedder to install an entropy source
+// using v8::V8::SetEntropySource(), which provides reasonable entropy, see:
+// https://code.google.com/p/v8/issues/detail?id=2905
 // This class is neither reentrant nor threadsafe.
 
 class RandomNumberGenerator V8_FINAL {