#define sha256msg1(m,a) \	// final SHA256MSG1 for middle rounds that require it
	SHA256MSG1		m, a

#define vmov(a,b) \		// msg copy for all but rounds 12-15
	VMOVDQA		a, b

#define vmovrev(a,b) \		// reverse copy for rounds 12-15
	VMOVDQA		b, a

// sha rounds 0 to 11
// identical with the exception of the final msg op
// which is replaced with a nop for rounds where it is not needed