loop3:
	MOVW (DI), SI
	CMPW SI,R8
	JZ   partial_success3
	ADDQ $1,DI
	CMPQ DI,DX
	JB loop3
	JMP fail
partial_success3:
	MOVW 1(DI), SI
	CMPW SI,BX
	JZ success
	ADDQ $1,DI
	CMPQ DI,DX
	JB loop3
	JMP fail
_4_or_more:
	CMPQ AX, $4
	JA   _5_or_more
	MOVL (R8), R8
	LEAQ -3(DI)(DX*1), DX
loop4:
	MOVL (DI), SI
	CMPL SI,R8
	JZ   success
	ADDQ $1,DI
	CMPQ DI,DX
	JB loop4
	JMP fail

Robert Griesemer <******@****.***> 1606801129 -0800

Aaron Jacobs <******@****.***> 1435103412 +1000

}
func rewriteValue386splitload_Op386CMPBload(v *Value) bool {
	v_2 := v.Args[2]
	v_1 := v.Args[1]
	v_0 := v.Args[0]
	b := v.Block
	typ := &b.Func.Config.Types
	// match: (CMPBload {sym} [off] ptr x mem)
	// result: (CMPB (MOVBload {sym} [off] ptr mem) x)
	for {
		off := auxIntToInt32(v.AuxInt)
		sym := auxToSym(v.Aux)
		ptr := v_0
		x := v_1
		mem := v_2
		v.reset(Op386CMPB)
		v0 := b.NewValue0(v.Pos, Op386MOVBload, typ.UInt8)

	// Shuffle X0 around so that each byte contains
	// the character we're looking for.
	MOVD AX, X0
	PUNPCKLBW X0, X0
	PUNPCKLBW X0, X0
	PSHUFL $0, X0, X0

	CMPQ BX, $16
	JLT small

	MOVQ SI, DI

	CMPQ BX, $32
	JA avx2
sse:
	LEAQ	-16(SI)(BX*1), AX	// AX = address of last 16 bytes
	JMP	sseloopentry

	PCALIGN $16
sseloop:
	// Move the next 16-byte chunk of the data into X1.

	echo $pkg
	DIR=$GOROOT/src/$pkg
	go build -gcflags "$FLAGS2 -S" -o /dev/null $pkg &> $DIR/new.txt
done

echo
echo
echo "3) compare assembly files"
for pkg in `go list std`; do
	DIR=$GOROOT/src/$pkg

	if cmp $DIR/old.txt $DIR/new.txt &> /dev/null
	then rm $DIR/old.txt $DIR/new.txt
	else echo "==> $DIR"
	fi

// Copyright 2023 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 cmp provides types and functions related to comparing
// ordered values.
package cmp

// Ordered is a constraint that permits any ordered type: any type
// that supports the operators < <= >= >.
// If future releases of Go add new ordered types,

	TESTL	BX, BX
	JEQ	_0
	CMPL	BX, $2
	JBE	_1or2
	CMPL	BX, $4
	JB	_3
	JE	_4
	CMPL	BX, $8
	JBE	_5through8
	CMPL	BX, $16
	JBE	_9through16
#ifdef GO386_softfloat
	JMP	nosse2
#endif
	PXOR	X0, X0
	CMPL	BX, $32
	JBE	_17through32
	CMPL	BX, $64
	JBE	_33through64
	CMPL	BX, $128
	JBE	_65through128
	CMPL	BX, $256
	JBE	_129through256

loop:
	MOVOU	X0, 0(DI)

	POPL		AX

// LTYPE3 rimrem	{ outcode(int($1), &$2); }
	SUBB	$1, AX
	SUBB	$1, foo+4(SB)
	SUBB	BX, AX
	SUBB	BX, foo+4(SB)

// LTYPE4 remrim	{ outcode(int($1), &$2); }
	CMPB	AX, $1
	CMPB	foo+4(SB), $4
	CMPB	BX, AX
	CMPB	foo+4(SB), BX

// LTYPER nonrel	{ outcode(int($1), &$2); }
label:
	JC	label // JCS
	JC	-1(PC) // JCS -1(PC)

// LTYPEC spec3	{ outcode(int($1), &$2); }
	CALL	AX
	JCS	2(PC)

func si1c(c <-chan int64) {
	// amd64:"CMPQ\t.+, [$]256"
	// s390x:"CLGIJ\t[$]12, R[0-9]+, [$]255"
	for x := <-c; x >= 0 && x < 256; x = <-c {
	}
}

func si2c(c <-chan int32) {
	// amd64:"CMPL\t.+, [$]256"
	// s390x:"CLIJ\t[$]12, R[0-9]+, [$]255"
	for x := <-c; x >= 0 && x < 256; x = <-c {
	}
}

func si3c(c <-chan int16) {
	// amd64:"CMPW\t.+, [$]256"
	// s390x:"CLIJ\t[$]12, R[0-9]+, [$]255"