mirror of
https://github.com/golang/go.git
synced 2025-05-05 23:53:05 +00:00
Add rewrite rules to optimize math.Copysign() when the second argument is negative floating point constant. For example, math.Copysign(c, -2): The previous compile output is "AND $9223372036854775807, R0, R0; ORR $-9223372036854775808, R0, R0". The optimized compile output is "ORR $-9223372036854775808, R0, R0" Math package benchmark results. name old time/op new time/op delta Copysign-8 2.61ns ± 2% 2.49ns ± 0% -4.55% (p=0.000 n=10+10) Cos-8 43.0ns ± 0% 41.5ns ± 0% -3.49% (p=0.000 n=10+10) Cosh-8 98.6ns ± 0% 98.1ns ± 0% -0.51% (p=0.000 n=10+10) ExpGo-8 107ns ± 0% 105ns ± 0% -1.87% (p=0.000 n=10+10) Exp2Go-8 100ns ± 0% 100ns ± 0% +0.39% (p=0.000 n=10+8) Max-8 6.56ns ± 2% 6.45ns ± 1% -1.63% (p=0.002 n=10+10) Min-8 6.66ns ± 3% 6.47ns ± 2% -2.82% (p=0.006 n=10+10) Mod-8 107ns ± 1% 104ns ± 1% -2.72% (p=0.000 n=10+10) Frexp-8 11.5ns ± 1% 11.0ns ± 0% -4.56% (p=0.000 n=8+10) HypotGo-8 19.4ns ± 0% 19.4ns ± 0% +0.36% (p=0.019 n=10+10) Ilogb-8 8.63ns ± 0% 8.51ns ± 0% -1.36% (p=0.000 n=10+10) Jn-8 584ns ± 0% 585ns ± 0% +0.17% (p=0.000 n=7+8) Ldexp-8 13.8ns ± 0% 13.5ns ± 0% -2.17% (p=0.002 n=8+10) Logb-8 10.2ns ± 0% 9.9ns ± 0% -2.65% (p=0.000 n=10+7) Nextafter64-8 7.54ns ± 0% 7.51ns ± 0% -0.37% (p=0.000 n=10+10) Remainder-8 73.5ns ± 1% 70.4ns ± 1% -4.27% (p=0.000 n=10+10) SqrtGoLatency-8 79.6ns ± 0% 76.2ns ± 0% -4.30% (p=0.000 n=9+10) Yn-8 582ns ± 0% 579ns ± 0% -0.52% (p=0.000 n=10+10) Change-Id: I0c9cd1ea87435e7b8bab94b4e79e6e29785f25b1 Reviewed-on: https://go-review.googlesource.com/132915 Reviewed-by: Cherry Zhang <cherryyz@google.com> Run-TryBot: Cherry Zhang <cherryyz@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org>
148 lines
3.8 KiB
Go
148 lines
3.8 KiB
Go
// asmcheck
|
|
|
|
// Copyright 2018 The Go Authors. All rights reserved.
|
|
// Use of this source code is governed by a BSD-style
|
|
// license that can be found in the LICENSE file.
|
|
|
|
package codegen
|
|
|
|
import "math"
|
|
|
|
var sink64 [8]float64
|
|
|
|
func approx(x float64) {
|
|
// s390x:"FIDBR\t[$]6"
|
|
// arm64:"FRINTPD"
|
|
// ppc64le:"FRIP"
|
|
sink64[0] = math.Ceil(x)
|
|
|
|
// s390x:"FIDBR\t[$]7"
|
|
// arm64:"FRINTMD"
|
|
// ppc64le:"FRIM"
|
|
sink64[1] = math.Floor(x)
|
|
|
|
// s390x:"FIDBR\t[$]1"
|
|
// arm64:"FRINTAD"
|
|
// ppc64le:"FRIN"
|
|
sink64[2] = math.Round(x)
|
|
|
|
// s390x:"FIDBR\t[$]5"
|
|
// arm64:"FRINTZD"
|
|
// ppc64le:"FRIZ"
|
|
sink64[3] = math.Trunc(x)
|
|
|
|
// s390x:"FIDBR\t[$]4"
|
|
sink64[4] = math.RoundToEven(x)
|
|
}
|
|
|
|
func sqrt(x float64) float64 {
|
|
// amd64:"SQRTSD"
|
|
// 386/387:"FSQRT" 386/sse2:"SQRTSD"
|
|
// arm64:"FSQRTD"
|
|
// arm/7:"SQRTD"
|
|
// mips/hardfloat:"SQRTD" mips/softfloat:-"SQRTD"
|
|
// mips64/hardfloat:"SQRTD" mips64/softfloat:-"SQRTD"
|
|
return math.Sqrt(x)
|
|
}
|
|
|
|
// Check that it's using integer registers
|
|
func abs(x, y float64) {
|
|
// amd64:"BTRQ\t[$]63"
|
|
// s390x:"LPDFR\t",-"MOVD\t" (no integer load/store)
|
|
// ppc64le:"FABS\t"
|
|
sink64[0] = math.Abs(x)
|
|
|
|
// amd64:"BTRQ\t[$]63","PXOR" (TODO: this should be BTSQ)
|
|
// s390x:"LNDFR\t",-"MOVD\t" (no integer load/store)
|
|
// ppc64le:"FNABS\t"
|
|
sink64[1] = -math.Abs(y)
|
|
}
|
|
|
|
// Check that it's using integer registers
|
|
func abs32(x float32) float32 {
|
|
// s390x:"LPDFR",-"LDEBR",-"LEDBR" (no float64 conversion)
|
|
return float32(math.Abs(float64(x)))
|
|
}
|
|
|
|
// Check that it's using integer registers
|
|
func copysign(a, b, c float64) {
|
|
// amd64:"BTRQ\t[$]63","SHRQ\t[$]63","SHLQ\t[$]63","ORQ"
|
|
// s390x:"CPSDR",-"MOVD" (no integer load/store)
|
|
// ppc64le:"FCPSGN"
|
|
sink64[0] = math.Copysign(a, b)
|
|
|
|
// amd64:"BTSQ\t[$]63"
|
|
// s390x:"LNDFR\t",-"MOVD\t" (no integer load/store)
|
|
// ppc64le:"FCPSGN"
|
|
// arm64:"ORR\t[$]-9223372036854775808"
|
|
sink64[1] = math.Copysign(c, -1)
|
|
|
|
// Like math.Copysign(c, -1), but with integer operations. Useful
|
|
// for platforms that have a copysign opcode to see if it's detected.
|
|
// s390x:"LNDFR\t",-"MOVD\t" (no integer load/store)
|
|
sink64[2] = math.Float64frombits(math.Float64bits(a) | 1<<63)
|
|
|
|
// amd64:-"SHLQ\t[$]1",-"SHRQ\t[$]1","SHRQ\t[$]63","SHLQ\t[$]63","ORQ"
|
|
// s390x:"CPSDR\t",-"MOVD\t" (no integer load/store)
|
|
// ppc64le:"FCPSGN"
|
|
sink64[3] = math.Copysign(-1, c)
|
|
}
|
|
|
|
func fromFloat64(f64 float64) uint64 {
|
|
// amd64:"MOVQ\tX.*, [^X].*"
|
|
return math.Float64bits(f64+1) + 1
|
|
}
|
|
|
|
func fromFloat32(f32 float32) uint32 {
|
|
// amd64:"MOVL\tX.*, [^X].*"
|
|
return math.Float32bits(f32+1) + 1
|
|
}
|
|
|
|
func toFloat64(u64 uint64) float64 {
|
|
// amd64:"MOVQ\t[^X].*, X.*"
|
|
return math.Float64frombits(u64+1) + 1
|
|
}
|
|
|
|
func toFloat32(u32 uint32) float32 {
|
|
// amd64:"MOVL\t[^X].*, X.*"
|
|
return math.Float32frombits(u32+1) + 1
|
|
}
|
|
|
|
// Test that comparisons with constants converted to float
|
|
// are evaluated at compile-time
|
|
|
|
func constantCheck64() bool {
|
|
// amd64:"MOVB\t[$]0",-"FCMP",-"MOVB\t[$]1"
|
|
// s390x:"MOV(B|BZ|D)\t[$]0,",-"FCMPU",-"MOV(B|BZ|D)\t[$]1,"
|
|
return 0.5 == float64(uint32(1)) || 1.5 > float64(uint64(1<<63)) || math.NaN() == math.NaN()
|
|
}
|
|
|
|
func constantCheck32() bool {
|
|
// amd64:"MOVB\t[$]1",-"FCMP",-"MOVB\t[$]0"
|
|
// s390x:"MOV(B|BZ|D)\t[$]1,",-"FCMPU",-"MOV(B|BZ|D)\t[$]0,"
|
|
return float32(0.5) <= float32(int64(1)) && float32(1.5) >= float32(int32(-1<<31)) && float32(math.NaN()) != float32(math.NaN())
|
|
}
|
|
|
|
// Test that integer constants are converted to floating point constants
|
|
// at compile-time
|
|
|
|
func constantConvert32(x float32) float32 {
|
|
// amd64:"MOVSS\t[$]f32.3f800000\\(SB\\)"
|
|
// s390x:"FMOVS\t[$]f32.3f800000\\(SB\\)"
|
|
// ppc64le:"FMOVS\t[$]f32.3f800000\\(SB\\)"
|
|
if x > math.Float32frombits(0x3f800000) {
|
|
return -x
|
|
}
|
|
return x
|
|
}
|
|
|
|
func constantConvertInt32(x uint32) uint32 {
|
|
// amd64:-"MOVSS"
|
|
// s390x:-"FMOVS"
|
|
// ppc64le:-"FMOVS"
|
|
if x > math.Float32bits(1) {
|
|
return -x
|
|
}
|
|
return x
|
|
}
|