[builtins] Introduce proper Float64Log2 and Float64Log10 operators.

src/builtins.h

Index: src/builtins.h
diff --git a/src/builtins.h b/src/builtins.h
index 7cafa0fcbc7653cdf966ca4765cc77c6484786b7..67ffabfe3e30a5cb655f6f938442a06c0bd9e0ae 100644
--- a/src/builtins.h
+++ b/src/builtins.h
@@ -323,6 +323,8 @@ inline bool operator&(BuiltinExtraArguments lhs, BuiltinExtraArguments rhs) {
   V(MathFloor, 2)                     \
   V(MathLog, 2)                       \
   V(MathLog1p, 2)                     \
+  V(MathLog2, 2)                      \
+  V(MathLog10, 2)                     \
   V(MathRound, 2)                     \
   V(MathSqrt, 2)                      \
   V(MathTrunc, 2)                     \
@@ -638,6 +640,10 @@ class Builtins {
   static void Generate_MathLog(CodeStubAssembler* assembler);
   // ES6 section 20.2.2.21 Math.log ( x )
   static void Generate_MathLog1p(CodeStubAssembler* assembler);
+
+  static void Generate_MathLog2(CodeStubAssembler* assembler);
+  static void Generate_MathLog10(CodeStubAssembler* assembler);
+
   enum class MathMaxMinKind { kMax, kMin };
   static void Generate_MathMaxMin(MacroAssembler* masm, MathMaxMinKind kind);
   // ES6 section 20.2.2.24 Math.max ( value1, value2 , ...values )