mirror of
https://github.com/golang/go
synced 2024-11-27 04:31:21 -07:00
3b321a9d12
This CL add runtime.memmove inlining for AMD64 and ARM64. According to ssa dump from testcases generic rules can't inline memmomve properly due to one of the arguments is Phi operation. But this Phi op will be optimized out by later optimization stages. As a result memmove can be inlined during arch-specific rules. The commit add new optimization rules to arch-specific rules that can inline runtime.memmove if it possible during lowering stage. Optimization fires 5 times in Go source-code using regabi. Fixes #41662 Change-Id: Iaffaf4c482d068b5f0683d141863892202cc8824 Reviewed-on: https://go-review.googlesource.com/c/go/+/289151 Reviewed-by: Keith Randall <khr@golang.org> Run-TryBot: Keith Randall <khr@golang.org> TryBot-Result: Go Bot <gobot@golang.org> Trust: David Chase <drchase@google.com>
161 lines
3.2 KiB
Go
161 lines
3.2 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 "runtime"
|
|
|
|
// Check small copies are replaced with moves.
|
|
|
|
func movesmall4() {
|
|
x := [...]byte{1, 2, 3, 4}
|
|
// 386:-".*memmove"
|
|
// amd64:-".*memmove"
|
|
// arm:-".*memmove"
|
|
// arm64:-".*memmove"
|
|
// ppc64:-".*memmove"
|
|
// ppc64le:-".*memmove"
|
|
copy(x[1:], x[:])
|
|
}
|
|
|
|
func movesmall7() {
|
|
x := [...]byte{1, 2, 3, 4, 5, 6, 7}
|
|
// 386:-".*memmove"
|
|
// amd64:-".*memmove"
|
|
// arm64:-".*memmove"
|
|
// ppc64:-".*memmove"
|
|
// ppc64le:-".*memmove"
|
|
copy(x[1:], x[:])
|
|
}
|
|
|
|
func movesmall16() {
|
|
x := [...]byte{1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16}
|
|
// amd64:-".*memmove"
|
|
// ppc64:".*memmove"
|
|
// ppc64le:".*memmove"
|
|
copy(x[1:], x[:])
|
|
}
|
|
|
|
var x [256]byte
|
|
|
|
// Check that large disjoint copies are replaced with moves.
|
|
|
|
func moveDisjointStack32() {
|
|
var s [32]byte
|
|
// ppc64:-".*memmove"
|
|
// ppc64le:-".*memmove"
|
|
// ppc64le/power8:"LXVD2X",-"ADD",-"BC"
|
|
// ppc64le/power9:"LXV",-"LXVD2X",-"ADD",-"BC"
|
|
copy(s[:], x[:32])
|
|
runtime.KeepAlive(&s)
|
|
}
|
|
|
|
func moveDisjointStack64() {
|
|
var s [96]byte
|
|
// ppc64:-".*memmove"
|
|
// ppc64le:-".*memmove"
|
|
// ppc64le/power8:"LXVD2X","ADD","BC"
|
|
// ppc64le/power9:"LXV",-"LXVD2X",-"ADD",-"BC"
|
|
copy(s[:], x[:96])
|
|
runtime.KeepAlive(&s)
|
|
}
|
|
|
|
func moveDisjointStack() {
|
|
var s [256]byte
|
|
// s390x:-".*memmove"
|
|
// amd64:-".*memmove"
|
|
// ppc64:-".*memmove"
|
|
// ppc64le:-".*memmove"
|
|
// ppc64le/power8:"LXVD2X"
|
|
// ppc64le/power9:"LXV",-"LXVD2X"
|
|
copy(s[:], x[:])
|
|
runtime.KeepAlive(&s)
|
|
}
|
|
|
|
func moveDisjointArg(b *[256]byte) {
|
|
var s [256]byte
|
|
// s390x:-".*memmove"
|
|
// amd64:-".*memmove"
|
|
// ppc64:-".*memmove"
|
|
// ppc64le:-".*memmove"
|
|
// ppc64le/power8:"LXVD2X"
|
|
// ppc64le/power9:"LXV",-"LXVD2X"
|
|
copy(s[:], b[:])
|
|
runtime.KeepAlive(&s)
|
|
}
|
|
|
|
func moveDisjointNoOverlap(a *[256]byte) {
|
|
// s390x:-".*memmove"
|
|
// amd64:-".*memmove"
|
|
// ppc64:-".*memmove"
|
|
// ppc64le:-".*memmove"
|
|
// ppc64le/power8:"LXVD2X"
|
|
// ppc64le/power9:"LXV",-"LXVD2X"
|
|
copy(a[:], a[128:])
|
|
}
|
|
|
|
// Check arch-specific memmove lowering. See issue 41662 fot details
|
|
|
|
func moveArchLowering1(b []byte, x *[1]byte) {
|
|
_ = b[1]
|
|
// amd64:-".*memmove"
|
|
// arm64:-".*memmove"
|
|
copy(b, x[:])
|
|
}
|
|
|
|
func moveArchLowering2(b []byte, x *[2]byte) {
|
|
_ = b[2]
|
|
// amd64:-".*memmove"
|
|
// arm64:-".*memmove"
|
|
copy(b, x[:])
|
|
}
|
|
|
|
func moveArchLowering4(b []byte, x *[4]byte) {
|
|
_ = b[4]
|
|
// amd64:-".*memmove"
|
|
// arm64:-".*memmove"
|
|
copy(b, x[:])
|
|
}
|
|
|
|
func moveArchLowering8(b []byte, x *[8]byte) {
|
|
_ = b[8]
|
|
// amd64:-".*memmove"
|
|
// arm64:-".*memmove"
|
|
copy(b, x[:])
|
|
}
|
|
|
|
func moveArchLowering16(b []byte, x *[16]byte) {
|
|
_ = b[16]
|
|
// amd64:-".*memmove"
|
|
copy(b, x[:])
|
|
}
|
|
|
|
// Check that no branches are generated when the pointers are [not] equal.
|
|
|
|
func ptrEqual() {
|
|
// amd64:-"JEQ",-"JNE"
|
|
// ppc64:-"BEQ",-"BNE"
|
|
// ppc64le:-"BEQ",-"BNE"
|
|
// s390x:-"BEQ",-"BNE"
|
|
copy(x[:], x[:])
|
|
}
|
|
|
|
func ptrOneOffset() {
|
|
// amd64:-"JEQ",-"JNE"
|
|
// ppc64:-"BEQ",-"BNE"
|
|
// ppc64le:-"BEQ",-"BNE"
|
|
// s390x:-"BEQ",-"BNE"
|
|
copy(x[1:], x[:])
|
|
}
|
|
|
|
func ptrBothOffset() {
|
|
// amd64:-"JEQ",-"JNE"
|
|
// ppc64:-"BEQ",-"BNE"
|
|
// ppc64le:-"BEQ",-"BNE"
|
|
// s390x:-"BEQ",-"BNE"
|
|
copy(x[1:], x[2:])
|
|
}
|