Speculative fix for computing Math.pow(2, -1074) on win32 where...

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 4152 matching lines @@
   while (true) {
     if ((n & 1) != 0) p *= m;
     n >>= 1;
     if (n == 0) {
       if (y < 0) {
         // Unfortunately, we have to be careful when p has reached
         // infinity in the computation, because sometimes the higher
         // internal precision in the pow() implementation would have
         // given us a finite p. This happens very rarely.
         double result = 1.0 / p;
-        return (result == 0 && isinf(p)) ? pow(x, y) : result;
+        return (result == 0 && isinf(p))
+            ? pow(x, static_cast<double>(y))  // Avoid pow(double, int).
+            : result;
       } else {
         return p;
       }
     }
     m *= m;
   }
 }
 
 
 static Object* Runtime_Math_pow(Arguments args) {
@@ skipping 3315 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();
   }
 }
 
 
 } }  // namespace v8::internal