Fix a bug that Math.round() returns incorrect results for huge integers.

src/runtime.cc

 // Copyright 2006-2009 the V8 project authors. All rights reserved.
 // Redistribution and use in source and binary forms, with or without
 // modification, are permitted provided that the following conditions are
 // met:
 //
 //     * Redistributions of source code must retain the above copyright
 //       notice, this list of conditions and the following disclaimer.
 //     * Redistributions in binary form must reproduce the above
 //       copyright notice, this list of conditions and the following
 //       disclaimer in the documentation and/or other materials provided
@@ skipping 4644 matching lines @@
   }
 }
 
 
 static Object* Runtime_Math_round(Arguments args) {
   NoHandleAllocation ha;
   ASSERT(args.length() == 1);
 
   CONVERT_DOUBLE_CHECKED(x, args[0]);
   if (signbit(x) && x >= -0.5) return Heap::minus_zero_value();
-  return Heap::NumberFromDouble(floor(x + 0.5));
+  double integer = ceil(x);
+  return Heap::NumberFromDouble(integer - x > 0.5 ? integer - 1.0 : integer);
 }
 
 
 static Object* Runtime_Math_sin(Arguments args) {
   NoHandleAllocation ha;
   ASSERT(args.length() == 1);
 
   CONVERT_DOUBLE_CHECKED(x, args[0]);
   return TranscendentalCache::Get(TranscendentalCache::SIN, x);
 }
@@ skipping 3490 matching lines @@
   } else {
     // Handle last resort GC and make sure to allow future allocations
     // to grow the heap without causing GCs (if possible).
     Counters::gc_last_resort_from_js.Increment();
     Heap::CollectAllGarbage(false);
   }
 }
 
 
 } }  // namespace v8::internal