mirror of
https://github.com/golang/go
synced 2024-11-16 18:04:39 -07:00
9eb53ab9bc
Add SSA rules to intrinsify Mul/Mul64 (AMD64 and ARM64). SSA rules for other functions and architectures are left as a future optimization. Benchmark results on AMD64/ARM64 before and after SSA implementation are below. amd64 name old time/op new time/op delta Add-4 1.78ns ± 0% 1.85ns ±12% ~ (p=0.397 n=4+5) Add32-4 1.71ns ± 1% 1.70ns ± 0% ~ (p=0.683 n=5+5) Add64-4 1.80ns ± 2% 1.77ns ± 0% -1.22% (p=0.048 n=5+5) Sub-4 1.78ns ± 0% 1.78ns ± 0% ~ (all equal) Sub32-4 1.78ns ± 1% 1.78ns ± 0% ~ (p=1.000 n=5+5) Sub64-4 1.78ns ± 1% 1.78ns ± 0% ~ (p=0.968 n=5+4) Mul-4 11.5ns ± 1% 1.8ns ± 2% -84.39% (p=0.008 n=5+5) Mul32-4 1.39ns ± 0% 1.38ns ± 3% ~ (p=0.175 n=5+5) Mul64-4 6.85ns ± 1% 1.78ns ± 1% -73.97% (p=0.008 n=5+5) Div-4 57.1ns ± 1% 56.7ns ± 0% ~ (p=0.087 n=5+5) Div32-4 18.0ns ± 0% 18.0ns ± 0% ~ (all equal) Div64-4 56.4ns ±10% 53.6ns ± 1% ~ (p=0.071 n=5+5) arm64 name old time/op new time/op delta Add-96 5.51ns ± 0% 5.51ns ± 0% ~ (all equal) Add32-96 5.51ns ± 0% 5.51ns ± 0% ~ (all equal) Add64-96 5.52ns ± 0% 5.51ns ± 0% ~ (p=0.444 n=5+5) Sub-96 5.51ns ± 0% 5.51ns ± 0% ~ (all equal) Sub32-96 5.51ns ± 0% 5.51ns ± 0% ~ (all equal) Sub64-96 5.51ns ± 0% 5.51ns ± 0% ~ (all equal) Mul-96 34.6ns ± 0% 5.0ns ± 0% -85.52% (p=0.008 n=5+5) Mul32-96 4.51ns ± 0% 4.51ns ± 0% ~ (all equal) Mul64-96 21.1ns ± 0% 5.0ns ± 0% -76.26% (p=0.008 n=5+5) Div-96 64.7ns ± 0% 64.7ns ± 0% ~ (all equal) Div32-96 17.0ns ± 0% 17.0ns ± 0% ~ (all equal) Div64-96 53.1ns ± 0% 53.1ns ± 0% ~ (all equal) Updates #24813 Change-Id: I9bda6d2102f65cae3d436a2087b47ed8bafeb068 Reviewed-on: https://go-review.googlesource.com/129415 Run-TryBot: Keith Randall <khr@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
321 lines
5.9 KiB
Go
321 lines
5.9 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/bits"
|
|
|
|
// ----------------------- //
|
|
// bits.LeadingZeros //
|
|
// ----------------------- //
|
|
|
|
func LeadingZeros(n uint) int {
|
|
// amd64:"BSRQ"
|
|
// s390x:"FLOGR"
|
|
// arm:"CLZ" arm64:"CLZ"
|
|
// mips:"CLZ"
|
|
return bits.LeadingZeros(n)
|
|
}
|
|
|
|
func LeadingZeros64(n uint64) int {
|
|
// amd64:"BSRQ"
|
|
// s390x:"FLOGR"
|
|
// arm:"CLZ" arm64:"CLZ"
|
|
// mips:"CLZ"
|
|
return bits.LeadingZeros64(n)
|
|
}
|
|
|
|
func LeadingZeros32(n uint32) int {
|
|
// amd64:"BSRQ","LEAQ",-"CMOVQEQ"
|
|
// s390x:"FLOGR"
|
|
// arm:"CLZ" arm64:"CLZ"
|
|
// mips:"CLZ"
|
|
return bits.LeadingZeros32(n)
|
|
}
|
|
|
|
func LeadingZeros16(n uint16) int {
|
|
// amd64:"BSRL","LEAL",-"CMOVQEQ"
|
|
// s390x:"FLOGR"
|
|
// arm:"CLZ" arm64:"CLZ"
|
|
// mips:"CLZ"
|
|
return bits.LeadingZeros16(n)
|
|
}
|
|
|
|
func LeadingZeros8(n uint8) int {
|
|
// amd64:"BSRL","LEAL",-"CMOVQEQ"
|
|
// s390x:"FLOGR"
|
|
// arm:"CLZ" arm64:"CLZ"
|
|
// mips:"CLZ"
|
|
return bits.LeadingZeros8(n)
|
|
}
|
|
|
|
// --------------- //
|
|
// bits.Len* //
|
|
// --------------- //
|
|
|
|
func Len(n uint) int {
|
|
// amd64:"BSRQ"
|
|
// s390x:"FLOGR"
|
|
// arm:"CLZ" arm64:"CLZ"
|
|
// mips:"CLZ"
|
|
return bits.Len(n)
|
|
}
|
|
|
|
func Len64(n uint64) int {
|
|
// amd64:"BSRQ"
|
|
// s390x:"FLOGR"
|
|
// arm:"CLZ" arm64:"CLZ"
|
|
// mips:"CLZ"
|
|
return bits.Len64(n)
|
|
}
|
|
|
|
func Len32(n uint32) int {
|
|
// amd64:"BSRQ","LEAQ",-"CMOVQEQ"
|
|
// s390x:"FLOGR"
|
|
// arm:"CLZ" arm64:"CLZ"
|
|
// mips:"CLZ"
|
|
return bits.Len32(n)
|
|
}
|
|
|
|
func Len16(n uint16) int {
|
|
// amd64:"BSRL","LEAL",-"CMOVQEQ"
|
|
// s390x:"FLOGR"
|
|
// arm:"CLZ" arm64:"CLZ"
|
|
// mips:"CLZ"
|
|
return bits.Len16(n)
|
|
}
|
|
|
|
func Len8(n uint8) int {
|
|
// amd64:"BSRL","LEAL",-"CMOVQEQ"
|
|
// s390x:"FLOGR"
|
|
// arm:"CLZ" arm64:"CLZ"
|
|
// mips:"CLZ"
|
|
return bits.Len8(n)
|
|
}
|
|
|
|
// -------------------- //
|
|
// bits.OnesCount //
|
|
// -------------------- //
|
|
|
|
func OnesCount(n uint) int {
|
|
// amd64:"POPCNTQ",".*support_popcnt"
|
|
// arm64:"VCNT","VUADDLV"
|
|
// s390x:"POPCNT"
|
|
return bits.OnesCount(n)
|
|
}
|
|
|
|
func OnesCount64(n uint64) int {
|
|
// amd64:"POPCNTQ",".*support_popcnt"
|
|
// arm64:"VCNT","VUADDLV"
|
|
// s390x:"POPCNT"
|
|
return bits.OnesCount64(n)
|
|
}
|
|
|
|
func OnesCount32(n uint32) int {
|
|
// amd64:"POPCNTL",".*support_popcnt"
|
|
// arm64:"VCNT","VUADDLV"
|
|
// s390x:"POPCNT"
|
|
return bits.OnesCount32(n)
|
|
}
|
|
|
|
func OnesCount16(n uint16) int {
|
|
// amd64:"POPCNTL",".*support_popcnt"
|
|
// arm64:"VCNT","VUADDLV"
|
|
// s390x:"POPCNT"
|
|
return bits.OnesCount16(n)
|
|
}
|
|
|
|
func OnesCount8(n uint8) int {
|
|
// s390x:"POPCNT"
|
|
return bits.OnesCount8(n)
|
|
}
|
|
|
|
// ----------------------- //
|
|
// bits.ReverseBytes //
|
|
// ----------------------- //
|
|
|
|
func ReverseBytes(n uint) uint {
|
|
// amd64:"BSWAPQ"
|
|
// s390x:"MOVDBR"
|
|
// arm64:"REV"
|
|
return bits.ReverseBytes(n)
|
|
}
|
|
|
|
func ReverseBytes64(n uint64) uint64 {
|
|
// amd64:"BSWAPQ"
|
|
// s390x:"MOVDBR"
|
|
// arm64:"REV"
|
|
return bits.ReverseBytes64(n)
|
|
}
|
|
|
|
func ReverseBytes32(n uint32) uint32 {
|
|
// amd64:"BSWAPL"
|
|
// s390x:"MOVWBR"
|
|
// arm64:"REVW"
|
|
return bits.ReverseBytes32(n)
|
|
}
|
|
|
|
func ReverseBytes16(n uint16) uint16 {
|
|
// amd64:"ROLW"
|
|
return bits.ReverseBytes16(n)
|
|
}
|
|
|
|
// --------------------- //
|
|
// bits.RotateLeft //
|
|
// --------------------- //
|
|
|
|
func RotateLeft64(n uint64) uint64 {
|
|
// amd64:"ROLQ"
|
|
// arm64:"ROR"
|
|
// ppc64:"ROTL"
|
|
// s390x:"RLLG"
|
|
return bits.RotateLeft64(n, 37)
|
|
}
|
|
|
|
func RotateLeft32(n uint32) uint32 {
|
|
// amd64:"ROLL" 386:"ROLL"
|
|
// arm64:"RORW"
|
|
// ppc64:"ROTLW"
|
|
// s390x:"RLL"
|
|
return bits.RotateLeft32(n, 9)
|
|
}
|
|
|
|
func RotateLeft16(n uint16) uint16 {
|
|
// amd64:"ROLW" 386:"ROLW"
|
|
return bits.RotateLeft16(n, 5)
|
|
}
|
|
|
|
func RotateLeft8(n uint8) uint8 {
|
|
// amd64:"ROLB" 386:"ROLB"
|
|
return bits.RotateLeft8(n, 5)
|
|
}
|
|
|
|
func RotateLeftVariable(n uint, m int) uint {
|
|
// amd64:"ROLQ"
|
|
// arm64:"ROR"
|
|
// ppc64:"ROTL"
|
|
// s390x:"RLLG"
|
|
return bits.RotateLeft(n, m)
|
|
}
|
|
|
|
func RotateLeftVariable64(n uint64, m int) uint64 {
|
|
// amd64:"ROLQ"
|
|
// arm64:"ROR"
|
|
// ppc64:"ROTL"
|
|
// s390x:"RLLG"
|
|
return bits.RotateLeft64(n, m)
|
|
}
|
|
|
|
func RotateLeftVariable32(n uint32, m int) uint32 {
|
|
// amd64:"ROLL"
|
|
// arm64:"RORW"
|
|
// ppc64:"ROTLW"
|
|
// s390x:"RLL"
|
|
return bits.RotateLeft32(n, m)
|
|
}
|
|
|
|
// ------------------------ //
|
|
// bits.TrailingZeros //
|
|
// ------------------------ //
|
|
|
|
func TrailingZeros(n uint) int {
|
|
// amd64:"BSFQ","MOVL\t\\$64","CMOVQEQ"
|
|
// s390x:"FLOGR"
|
|
return bits.TrailingZeros(n)
|
|
}
|
|
|
|
func TrailingZeros64(n uint64) int {
|
|
// amd64:"BSFQ","MOVL\t\\$64","CMOVQEQ"
|
|
// s390x:"FLOGR"
|
|
return bits.TrailingZeros64(n)
|
|
}
|
|
|
|
func TrailingZeros32(n uint32) int {
|
|
// amd64:"BTSQ\\t\\$32","BSFQ"
|
|
// s390x:"FLOGR","MOVWZ"
|
|
return bits.TrailingZeros32(n)
|
|
}
|
|
|
|
func TrailingZeros16(n uint16) int {
|
|
// amd64:"BSFL","BTSL\\t\\$16"
|
|
// s390x:"FLOGR","OR\t\\$65536"
|
|
return bits.TrailingZeros16(n)
|
|
}
|
|
|
|
func TrailingZeros8(n uint8) int {
|
|
// amd64:"BSFL","BTSL\\t\\$8"
|
|
// s390x:"FLOGR","OR\t\\$256"
|
|
return bits.TrailingZeros8(n)
|
|
}
|
|
|
|
// IterateBitsNN checks special handling of TrailingZerosNN when the input is known to be non-zero.
|
|
|
|
func IterateBits(n uint) int {
|
|
i := 0
|
|
for n != 0 {
|
|
// amd64:"BSFQ",-"CMOVEQ"
|
|
i += bits.TrailingZeros(n)
|
|
n &= n - 1
|
|
}
|
|
return i
|
|
}
|
|
|
|
func IterateBits64(n uint64) int {
|
|
i := 0
|
|
for n != 0 {
|
|
// amd64:"BSFQ",-"CMOVEQ"
|
|
i += bits.TrailingZeros64(n)
|
|
n &= n - 1
|
|
}
|
|
return i
|
|
}
|
|
|
|
func IterateBits32(n uint32) int {
|
|
i := 0
|
|
for n != 0 {
|
|
// amd64:"BSFL",-"BTSQ"
|
|
i += bits.TrailingZeros32(n)
|
|
n &= n - 1
|
|
}
|
|
return i
|
|
}
|
|
|
|
func IterateBits16(n uint16) int {
|
|
i := 0
|
|
for n != 0 {
|
|
// amd64:"BSFL",-"BTSL"
|
|
i += bits.TrailingZeros16(n)
|
|
n &= n - 1
|
|
}
|
|
return i
|
|
}
|
|
|
|
func IterateBits8(n uint8) int {
|
|
i := 0
|
|
for n != 0 {
|
|
// amd64:"BSFL",-"BTSL"
|
|
i += bits.TrailingZeros8(n)
|
|
n &= n - 1
|
|
}
|
|
return i
|
|
}
|
|
|
|
// --------------- //
|
|
// bits.Mul* //
|
|
// --------------- //
|
|
|
|
func Mul(x, y uint) (hi, lo uint) {
|
|
// amd64:"MULQ"
|
|
// arm64:"UMULH","MUL"
|
|
return bits.Mul(x, y)
|
|
}
|
|
|
|
func Mul64(x, y uint64) (hi, lo uint64) {
|
|
// amd64:"MULQ"
|
|
// arm64:"UMULH","MUL"
|
|
return bits.Mul64(x, y)
|
|
}
|