[turbofan] Implement rounding of floats on x64 and ia32 without sse4.1.

src/assembler.h

 // Copyright (c) 1994-2006 Sun Microsystems Inc.
 // 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.
 //
@@ skipping 947 matching lines @@
   static ExternalReference scheduled_exception_address(Isolate* isolate);
   static ExternalReference address_of_pending_message_obj(Isolate* isolate);
 
   // Static variables containing common double constants.
   static ExternalReference address_of_min_int();
   static ExternalReference address_of_one_half();
   static ExternalReference address_of_minus_one_half();
   static ExternalReference address_of_negative_infinity();
   static ExternalReference address_of_the_hole_nan();
   static ExternalReference address_of_uint32_bias();
+  static ExternalReference address_of_rounding_limit();
 
   static ExternalReference math_log_double_function(Isolate* isolate);
 
   static ExternalReference math_exp_constants(int constant_index);
   static ExternalReference math_exp_log_table();
 
   static ExternalReference page_flags(Page* page);
 
   static ExternalReference ForDeoptEntry(Address entry);
 
@@ skipping 301 matching lines @@
     std::vector<ConstantPoolEntry> shared_entries;
   };
 
   Label emitted_label_;  // Records pc_offset of emitted pool
   PerTypeEntryInfo info_[ConstantPoolEntry::NUMBER_OF_TYPES];
 };
 
 }  // namespace internal
 }  // namespace v8
 #endif  // V8_ASSEMBLER_H_