qbit/go
1
0
Fork 0
mirror of https://github.com/golang/go synced 2024-09-29 21:34:28 -06:00
Code Releases Activity
cb65c8d58a
go/test/codegen/shift.go
Paul E. Murphy c3c6fbf314 cmd/compile: combine more 32 bit shift and mask operations on ppc64 
Combine (AND m (SRWconst x)) or (SRWconst (AND m x)) when mask m is
and the shift value produce constant which can be encoded into an
RLWINM instruction.

Combine (CLRLSLDI (SRWconst x)) if the combining of the underling rotate
masks produces a constant which can be encoded into RLWINM.

Likewise for (SLDconst (SRWconst x)) and (CLRLSDI (RLWINM x)).

Combine rotate word + and operations which can be encoded as a single
RLWINM/RLWNM instruction.

The most notable performance improvements arise from the crypto
benchmarks below (GOARCH=power8 on a ppc64le/linux):

pkg:golang.org/x/crypto/blowfish goos:linux goarch:ppc64le
ExpandKeyWithSalt                               52.2µs ± 0%    47.5µs ± 0%  -8.88%
ExpandKey                                       44.4µs ± 0%    40.3µs ± 0%  -9.15%

pkg:golang.org/x/crypto/ssh/internal/bcrypt_pbkdf goos:linux goarch:ppc64le
Key                                             57.6ms ± 0%    52.3ms ± 0%  -9.13%

pkg:golang.org/x/crypto/bcrypt goos:linux goarch:ppc64le
Equal                                           90.9ms ± 0%    82.6ms ± 0%  -9.13%
DefaultCost                                     91.0ms ± 0%    82.7ms ± 0%  -9.12%

Change-Id: I59a0ca29face38f4ab46e37124c32906f216c4ce
Reviewed-on: https://go-review.googlesource.com/c/go/+/260798
Run-TryBot: Carlos Eduardo Seo <carlos.seo@linaro.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
Reviewed-by: Carlos Eduardo Seo <carlos.seo@linaro.com>
Trust: Lynn Boger <laboger@linux.vnet.ibm.com>
2020-10-27 18:33:20 +00:00

288 lines
7.3 KiB
Go
Raw Blame History

// 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
// ------------------ //
// masked shifts //
// ------------------ //
func lshMask64x64(v int64, s uint64) int64 {
// s390x:-".*AND",-".*MOVDGE"
// ppc64le:"ANDCC",-"ORN",-"ISEL"
// ppc64:"ANDCC",-"ORN",-"ISEL"
return v << (s & 63)
}
func rshMask64Ux64(v uint64, s uint64) uint64 {
// s390x:-".*AND",-".*MOVDGE"
// ppc64le:"ANDCC",-"ORN",-"ISEL"
// ppc64:"ANDCC",-"ORN",-"ISEL"
return v >> (s & 63)
}
func rshMask64x64(v int64, s uint64) int64 {
// s390x:-".*AND",-".*MOVDGE"
// ppc64le:"ANDCC",-ORN",-"ISEL"
// ppc64:"ANDCC",-"ORN",-"ISEL"
return v >> (s & 63)
}
func lshMask32x64(v int32, s uint64) int32 {
// s390x:-".*AND",-".*MOVDGE"
// ppc64le:"ISEL",-"ORN"
// ppc64:"ISEL",-"ORN"
return v << (s & 63)
}
func rshMask32Ux64(v uint32, s uint64) uint32 {
// s390x:-".*AND",-".*MOVDGE"
// ppc64le:"ISEL",-"ORN"
// ppc64:"ISEL",-"ORN"
return v >> (s & 63)
}
func rshMask32x64(v int32, s uint64) int32 {
// s390x:-".*AND",-".*MOVDGE"
// ppc64le:"ISEL",-"ORN"
// ppc64:"ISEL",-"ORN"
return v >> (s & 63)
}
func lshMask64x32(v int64, s uint32) int64 {
// s390x:-".*AND",-".*MOVDGE"
// ppc64le:"ANDCC",-"ORN"
// ppc64:"ANDCC",-"ORN"
return v << (s & 63)
}
func rshMask64Ux32(v uint64, s uint32) uint64 {
// s390x:-".*AND",-".*MOVDGE"
// ppc64le:"ANDCC",-"ORN"
// ppc64:"ANDCC",-"ORN"
return v >> (s & 63)
}
func rshMask64x32(v int64, s uint32) int64 {
// s390x:-".*AND",-".*MOVDGE"
// ppc64le:"ANDCC",-"ORN",-"ISEL"
// ppc64:"ANDCC",-"ORN",-"ISEL"
return v >> (s & 63)
}
func lshMask64x32Ext(v int64, s int32) int64 {
// s390x:-".*AND",-".*MOVDGE"
// ppc64le:"ANDCC",-"ORN",-"ISEL"
// ppc64:"ANDCC",-"ORN",-"ISEL"
return v << uint(s&63)
}
func rshMask64Ux32Ext(v uint64, s int32) uint64 {
// s390x:-".*AND",-".*MOVDGE"
// ppc64le:"ANDCC",-"ORN",-"ISEL"
// ppc64:"ANDCC",-"ORN",-"ISEL"
return v >> uint(s&63)
}
func rshMask64x32Ext(v int64, s int32) int64 {
// s390x:-".*AND",-".*MOVDGE"
// ppc64le:"ANDCC",-"ORN",-"ISEL"
// ppc64:"ANDCC",-"ORN",-"ISEL"
return v >> uint(s&63)
}
// --------------- //
// signed shifts //
// --------------- //
// We do want to generate a test + panicshift for these cases.
func lshSigned(v8 int8, v16 int16, v32 int32, v64 int64, x int) {
// amd64:"TESTB"
_ = x << v8
// amd64:"TESTW"
_ = x << v16
// amd64:"TESTL"
_ = x << v32
// amd64:"TESTQ"
_ = x << v64
}
// We want to avoid generating a test + panicshift for these cases.
func lshSignedMasked(v8 int8, v16 int16, v32 int32, v64 int64, x int) {
// amd64:-"TESTB"
_ = x << (v8 & 7)
// amd64:-"TESTW"
_ = x << (v16 & 15)
// amd64:-"TESTL"
_ = x << (v32 & 31)
// amd64:-"TESTQ"
_ = x << (v64 & 63)
}
// ------------------ //
// bounded shifts //
// ------------------ //
func rshGuarded64(v int64, s uint) int64 {
if s < 64 {
// s390x:-".*AND",-".*MOVDGE" wasm:-"Select",-".*LtU"
return v >> s
}
panic("shift too large")
}
func rshGuarded64U(v uint64, s uint) uint64 {
if s < 64 {
// s390x:-".*AND",-".*MOVDGE" wasm:-"Select",-".*LtU"
return v >> s
}
panic("shift too large")
}
func lshGuarded64(v int64, s uint) int64 {
if s < 64 {
// s390x:-".*AND",-".*MOVDGE" wasm:-"Select",-".*LtU"
return v << s
}
panic("shift too large")
}
func checkUnneededTrunc(tab *[100000]uint32, d uint64, v uint32, h uint16, b byte) (uint32, uint64) {
// ppc64le:-".*RLWINM",-".*RLDICR",".*CLRLSLDI"
// ppc64:-".*RLWINM",-".*RLDICR",".*CLRLSLDI"
f := tab[byte(v)^b]
// ppc64le:-".*RLWINM",-".*RLDICR",".*CLRLSLDI"
// ppc64:-".*RLWINM",-".*RLDICR",".*CLRLSLDI"
f += tab[byte(v)&b]
// ppc64le:-".*RLWINM",-".*RLDICR",".*CLRLSLDI"
// ppc64:-".*RLWINM",-".*RLDICR",".*CLRLSLDI"
f += tab[byte(v)|b]
// ppc64le:-".*RLWINM",-".*RLDICR",".*CLRLSLDI"
// ppc64:-".*RLWINM",-".*RLDICR",".*CLRLSLDI"
f += tab[uint16(v)&h]
// ppc64le:-".*RLWINM",-".*RLDICR",".*CLRLSLDI"
// ppc64:-".*RLWINM",-".*RLDICR",".*CLRLSLDI"
f += tab[uint16(v)^h]
// ppc64le:-".*RLWINM",-".*RLDICR",".*CLRLSLDI"
// ppc64:-".*RLWINM",-".*RLDICR",".*CLRLSLDI"
f += tab[uint16(v)|h]
// ppc64le:-".*AND",-"RLDICR",".*CLRLSLDI"
// ppc64:-".*AND",-"RLDICR",".*CLRLSLDI"
f += tab[v&0xff]
// ppc64le:-".*AND",".*CLRLSLWI"
// ppc64:-".*AND",".*CLRLSLWI"
f += 2 * uint32(uint16(d))
// ppc64le:-".*AND",-"RLDICR",".*CLRLSLDI"
// ppc64:-".*AND",-"RLDICR",".*CLRLSLDI"
g := 2 * uint64(uint32(d))
return f, g
}
func checkCombinedShifts(v8 uint8, v16 uint16, v32 uint32, x32 int32, v64 uint64) (uint8, uint16, uint32, uint64, int64) {
// ppc64le:-"AND","CLRLSLWI"
// ppc64:-"AND","CLRLSLWI"
f := (v8 & 0xF) << 2
// ppc64le:"CLRLSLWI"
// ppc64:"CLRLSLWI"
f += byte(v16) << 3
// ppc64le:-"AND","CLRLSLWI"
// ppc64:-"AND","CLRLSLWI"
g := (v16 & 0xFF) << 3
// ppc64le:-"AND","CLRLSLWI"
// ppc64:-"AND","CLRLSLWI"
h := (v32 & 0xFFFFF) << 2
// ppc64le:"CLRLSLDI"
// ppc64:"CLRLSLDI"
i := (v64 & 0xFFFFFFFF) << 5
// ppc64le:-"CLRLSLDI"
// ppc64:-"CLRLSLDI"
i += (v64 & 0xFFFFFFF) << 38
// ppc64le/power9:-"CLRLSLDI"
// ppc64/power9:-"CLRLSLDI"
i += (v64 & 0xFFFF00) << 10
// ppc64le/power9:-"SLD","EXTSWSLI"
// ppc64/power9:-"SLD","EXTSWSLI"
j := int64(x32+32) * 8
return f, g, h, i, j
}
func checkWidenAfterShift(v int64, u uint64) (int64, uint64) {
// ppc64le:-".*MOVW"
f := int32(v >> 32)
// ppc64le:".*MOVW"
f += int32(v >> 31)
// ppc64le:-".*MOVH"
g := int16(v >> 48)
// ppc64le:".*MOVH"
g += int16(v >> 30)
// ppc64le:-".*MOVH"
g += int16(f >> 16)
// ppc64le:-".*MOVB"
h := int8(v >> 56)
// ppc64le:".*MOVB"
h += int8(v >> 28)
// ppc64le:-".*MOVB"
h += int8(f >> 24)
// ppc64le:".*MOVB"
h += int8(f >> 16)
return int64(h), uint64(g)
}
func checkShiftAndMask32(v []uint32) {
i := 0
// ppc64le: "RLWNM\t[$]24, R[0-9]+, [$]1044480, R[0-9]+"
// ppc64: "RLWNM\t[$]24, R[0-9]+, [$]1044480, R[0-9]+"
v[i] = (v[i] & 0xFF00000) >> 8
i++
// ppc64le: "RLWNM\t[$]26, R[0-9]+, [$]1020, R[0-9]+"
// ppc64: "RLWNM\t[$]26, R[0-9]+, [$]1020, R[0-9]+"
v[i] = (v[i] & 0xFF00) >> 6
i++
// ppc64le: "MOVW\tR0"
// ppc64: "MOVW\tR0"
v[i] = (v[i] & 0xFF) >> 8
i++
// ppc64le: "MOVW\tR0"
// ppc64: "MOVW\tR0"
v[i] = (v[i] & 0xF000000) >> 28
i++
// ppc64le: "RLWNM\t[$]26, R[0-9]+, [$]255, R[0-9]+"
// ppc64: "RLWNM\t[$]26, R[0-9]+, [$]255, R[0-9]+"
v[i] = (v[i] >> 6) & 0xFF
i++
// ppc64le: "RLWNM\t[$]26, R[0-9]+, [$]1044480, R[0-9]+"
// ppc64: "RLWNM\t[$]26, R[0-9]+, [$]1044480, R[0-9]+"
v[i] = (v[i] >> 6) & 0xFF000
i++
// ppc64le: "MOVW\tR0"
// ppc64: "MOVW\tR0"
v[i] = (v[i] >> 20) & 0xFF000
i++
// ppc64le: "MOVW\tR0"
// ppc64: "MOVW\tR0"
v[i] = (v[i] >> 24) & 0xFF00
i++
}
func checkMergedShifts32(a [256]uint32, b [256]uint64, u uint32, v uint32) {
//ppc64le: -"CLRLSLDI", "RLWNM\t[$]10, R[0-9]+, [$]1020, R[0-9]+"
//ppc64: -"CLRLSLDI", "RLWNM\t[$]10, R[0-9]+, [$]1020, R[0-9]+"
a[0] = a[uint8(v>>24)]
//ppc64le: -"CLRLSLDI", "RLWNM\t[$]11, R[0-9]+, [$]2040, R[0-9]+"
//ppc64: -"CLRLSLDI", "RLWNM\t[$]11, R[0-9]+, [$]2040, R[0-9]+"
b[0] = b[uint8(v>>24)]
//ppc64le: -"CLRLSLDI", "RLWNM\t[$]15, R[0-9]+, [$]2040, R[0-9]+"
//ppc64: -"CLRLSLDI", "RLWNM\t[$]15, R[0-9]+, [$]2040, R[0-9]+"
b[1] = b[(v>>20)&0xFF]
//ppc64le: -"SLD", "RLWNM\t[$]10, R[0-9]+, [$]1016, R[0-9]+"
//ppc64: -"SLD", "RLWNM\t[$]10, R[0-9]+, [$]1016, R[0-9]+"
b[2] = b[v>>25]
}
View Git Blame Copy Permalink