SSE accelerated uint128 left shift in Go

u128ls.go

package main

import "fmt"

func main() {
	var x, y uint64 = 0x123456789abcdef0, 0xfedcba9876543213
	fmt.Printf("Before shift: %016x%016x\n", x, y)

	//y <<= 8
	//x = x<<8 | y>>(64-8)
	y, x = shiftLeft128ASM(y, x)

	fmt.Printf("After shift:  %016x%016x\n", x, y)
}

//go:noescape
func shiftLeft128ASM(lo, hi uint64) (uint64, uint64)

u128ls_x86.s

//go:build i386 || amd64

#include "textflag.h"

TEXT ·shiftLeft128ASM(SB), NOSPLIT, $0-32
	MOVOA x+0(FP), X0
	PSLLDQ $1, X0
	MOVOA X0, x+16(FP)
	RET