// arg1 = address of src memory (in R16 aka arm64.REGRT1, changed as side effect)
		// arg2 = address of the last element of src
		// arg3 = mem
		// returns mem
		//	LDP.P	16(R16), (R25, Rtmp)
		//	STP.P	(R25, Rtmp), 16(R17)
		//	CMP	Rarg2, R16
		//	BLE	-3(PC)
		// Note: the-end-of-src may be not a valid pointer. it's a problem if it is spilled.
		// the-end-of-src - 16 is within the area to copy, ok to spill.
		{