tail:	// Count the last 0 - 31 bytes.
	CMP	R4, $16
	BLE	small_tail_p10
	LXV	0(R3), V0
	VCMPEQUB V0, V1, V0
	VCNTMBB	V0, $1, R14	// Sum the value of bit 0 of each byte of the compare into R14.
	SRD	$56, R14, R14	// The result of VCNTMBB is shifted. Unshift it.
	ADD	R14, R18, R18
	ADD	$16, R3, R3
	ANDCC	$15, R4, R4

small_tail_p10:
	SLD	$56, R4, R6
	LXVLL	R3, R6, V0
	VCMPEQUB V0, V1, V0