diff --git a/floatfuncs.j b/floatfuncs.j
index 8b5f5a6bd75a1..d9f5cf29c1292 100644
--- a/floatfuncs.j
+++ b/floatfuncs.j
@@ -35,6 +35,10 @@ cos(x::Float32) = boxf32(cos_float(unbox32(x)))
 ^(x::Float64, p::Int32) = boxf64(powi_float(unbox64(x),unbox32(p)))
 ^(x::Float32, p::Int32) = boxf32(powi_float(unbox32(x),unbox32(p)))
 
+sqrt(x::Real) = sqrt(float(x))
+sin(x::Real) = sin(float(x))
+cos(x::Real) = cos(float(x))
+
 num2hex(x::Float32) = uint2str(boxui32(unbox32(x)),16,16)
 num2hex(x::Float64) = uint2str(boxui64(unbox64(x)),16,16)
 
diff --git a/math_libm.j b/math_libm.j
index 25dffd46d9eef..b4bd9e04c1659 100644
--- a/math_libm.j
+++ b/math_libm.j
@@ -7,13 +7,13 @@ macro vectorize(f)
     end
 end
 
-for f = {:sin, :cos, :tan, :sinh, :cosh, :tanh, :asin, :acos, :atan, :log,
-         :log2, :log10, :log1p, :logb, :exp, :exp2, :expm1, :erf, :erfc,
-         :sqrt, :cbrt, :ceil, :floor, :nearbyint, :round, :rint, :trunc}
+for f = {:tan, :sinh, :cosh, :tanh, :asin, :acos, :atan, :log, :log2,
+         :log10, :log1p, :logb, :exp, :exp2, :expm1, :erf, :erfc,
+         :cbrt, :ceil, :floor, :nearbyint, :round, :rint, :trunc}
     @eval begin
         ($f)(x::Float64) = ccall(dlsym(libm,$string(f)), Float64, (Float64,), x)
         ($f)(x::Float32) = ccall(dlsym(libm,$strcat(string(f),"f")), Float32, (Float32,), x)
-        ($f)(x::Real) = ($f)(convert(Float64,x))
+        ($f)(x::Real) = ($f)(float(x))
         @vectorize $f
     end
 end
@@ -29,28 +29,24 @@ for f = {:lrint, :lround, :ilogb}
     end
 end
 
-abs(x::Float64) = ccall(dlsym(libm, :fabs), Float64, (Float64,), x)
+abs(x::Float64) = ccall(dlsym(libm, :fabs),  Float64, (Float64,), x)
 abs(x::Float32) = ccall(dlsym(libm, :fabsf), Float32, (Float32,), x)
 @vectorize abs
 
 for f = {:atan2, :pow, :fmod, :copysign, :hypot, :fmin, :fmax, :fdim}
     @eval begin
-        ($f)(x::Float64, y::Float64) = ccall(dlsym(libm,$string(f)),
-                                             Float64,
-                                             (Float64, Float64,),
-                                             x, y)
-        ($f)(x::Float32, y::Float32) = ccall(dlsym(libm,$strcat(string(f),"f")),
-                                             Float32,
-                                             (Float32, Float32),
-                                             x, y)
-        ($f)(x::Real, y::Real) = ($f)(convert(Float64,x),convert(Float64,y))
+        ($f)(x::Float64, y::Float64) =
+            ccall(dlsym(libm,$string(f)), Float64, (Float64, Float64,), x, y)
+        ($f)(x::Float32, y::Float32) =
+            ccall(dlsym(libm,$strcat(string(f),"f")), Float32, (Float32, Float32), x, y)
+        ($f)(x::Real, y::Real) = ($f)(float(x),float(y))
     end
 end
 
-ldexp(x::Float64,e::Int32) = ccall(dlsym(libm, :ldexp), Float64, (Float64,Int32), x, e)
+ldexp(x::Float64,e::Int32) = ccall(dlsym(libm, :ldexp),  Float64, (Float64,Int32), x, e)
 ldexp(x::Float32,e::Int32) = ccall(dlsym(libm, :ldexpf), Float32, (Float32,Int32), x, e)
 
-rand() = ccall(:rand_double, Float64, ())
-randf() = ccall(:rand_float, Float32, ())
-randui32() = ccall(:genrand_int32, Uint32, ())
-randn() = ccall(:randn, Float64, ())
+rand()     = ccall(:rand_double,   Float64, ())
+randf()    = ccall(:rand_float,    Float32, ())
+randui32() = ccall(:genrand_int32, Uint32,  ())
+randn()    = ccall(:randn,         Float64, ())