jdyfag

Author: christiangnrd

src/device/intrinsics/math.jl

@@ -2,24 +2,45 @@
 
 using Base: FastMath
 using Base.Math: throw_complex_domainerror
+import Core: Float32, Float16
 
 # TODO:
 # - wrap all intrinsics from include/metal/metal_math
 # - add support for vector types
 # - consider emitting LLVM intrinsics and lowering those in the back-end
 
+# Precompute irrationals for use on GPU
+# macro _const_convert(irr,T,r)
+#     :($T($irr, $r))
+# end
+for T in (:Float32, :Float16), irr in (:π, :ℯ), r in (:RoundUp, :RoundDown)
+    # newT = Symbol(:new,T)
+    @eval begin
+        # $newT(::typeof($irr), ::typeof($r)) = @_const_convert($irr, $T, $r)
+        # @device_override $T(::typeof($irr), ::typeof($r)) = @_const_convert($irr, $T, $r)
+        @device_override $T(::typeof($irr), ::typeof($r)) = Base.Rounding._convert_rounding($T, $irr, $r)
+    end
+end
+
+for T in (:Float32, :Float16), irr in (:π, :ℯ), r in (:RoundUp, :RoundDown)
+    @eval begin
+        @device_override $T(::typeof($irr), ::typeof($r)) = Base.Rounding._convert_rounding($T, $irr, $r)
+    end
+end
+
 ### Constants
 # π
-@device_override Core.Float32(::typeof(π), ::RoundingMode) = reinterpret(Float32, 0x40490fdb)        # 3.1415927f0 reinterpret(UInt32,Float32(reinterpret(Float64,0x400921FB60000000)))
-@device_override Core.Float32(::typeof(π), ::RoundingMode{:Down}) = reinterpret(Float32, 0x40490fda) # 3.1415925f0 prevfloat(reinterpret(UInt32,Float32(reinterpret(Float64,0x400921FB60000000))))
-@device_override Core.Float16(::typeof(π), ::RoundingMode{:Up}) = reinterpret(Float16, 0x4249)       # Float16(3.143)
-@device_override Core.Float16(::typeof(π), ::RoundingMode) = reinterpret(Float16, 0x4248)            # Float16(3.14)
+# @device_override Core.Float32(::typeof(π), ::RoundingMode) = reinterpret(Float32, 0x40490fdb)        # 3.1415927f0 reinterpret(UInt32,Float32(reinterpret(Float64,0x400921FB60000000)))
+# @device_override Core.Float32(::typeof(π), ::RoundingMode{:Down}) = reinterpret(Float32, 0x40490fda) # 3.1415925f0 prevfloat(reinterpret(UInt32,Float32(reinterpret(Float64,0x400921FB60000000))))
+
+# @device_override Core.Float16(::typeof(π), ::RoundingMode{:Up}) = reinterpret(Float16, 0x4249)       # Float16(3.143)
+# @device_override Core.Float16(::typeof(π), ::RoundingMode) = reinterpret(Float16, 0x4248)            # Float16(3.14)
 
 # ℯ
-@device_override Core.Float32(::typeof(ℯ), ::RoundingMode{:Up}) = reinterpret(Float32, 0x402df855)   # 2.718282f0 nextfloat(reinterpret(UInt32,Float32(reinterpret(Float64,0x4005BF0A80000000))))
-@device_override Core.Float32(::typeof(ℯ), ::RoundingMode) = reinterpret(Float32, 0x402df854)        # 2.7182817f0 reinterpret(UInt32,Float32(reinterpret(Float64,0x4005BF0A80000000)))
-@device_override Core.Float16(::typeof(ℯ), ::RoundingMode) = reinterpret(Float16, 0x4170)            # Float16(2.719)
-@device_override Core.Float16(::typeof(ℯ), ::RoundingMode{:Down}) = reinterpret(Float16, 0x416f)     # Float16(2.717)
+# @device_override Core.Float32(::typeof(ℯ), ::RoundingMode{:Up}) = reinterpret(Float32, 0x402df855)   # 2.718282f0 nextfloat(reinterpret(UInt32,Float32(reinterpret(Float64,0x4005BF0A80000000))))
+# @device_override Core.Float32(::typeof(ℯ), ::RoundingMode) = reinterpret(Float32, 0x402df854)        # 2.7182817f0 reinterpret(UInt32,Float32(reinterpret(Float64,0x4005BF0A80000000)))
+# @device_override Core.Float16(::typeof(ℯ), ::RoundingMode) = reinterpret(Float16, 0x4170)            # Float16(2.719)
+# @device_override Core.Float16(::typeof(ℯ), ::RoundingMode{:Down}) = reinterpret(Float16, 0x416f)     # Float16(2.717)
 
 ### Common Intrinsics
 @device_function clamp_fast(x::Float32, minval::Float32, maxval::Float32) = ccall("extern air.fast_clamp.f32", llvmcall, Cfloat, (Cfloat, Cfloat, Cfloat), x, minval, maxval)