{"VS30", "VS30"},
	{"VS31", "VS31"},
	{"VS32", "VS32"},
	{"VS33", "VS33"},
	{"VS34", "VS34"},
	{"VS35", "VS35"},
	{"VS36", "VS36"},
	{"VS37", "VS37"},
	{"VS38", "VS38"},
	{"VS39", "VS39"},
	{"VS40", "VS40"},
	{"VS41", "VS41"},
	{"VS42", "VS42"},
	{"VS43", "VS43"},
	{"VS44", "VS44"},
	{"VS45", "VS45"},
	{"VS46", "VS46"},
	{"VS47", "VS47"},
	{"VS48", "VS48"},
	{"VS49", "VS49"},
	{"VS50", "VS50"},

unpack the input values into vector registers VSLDOI $8, V0, V0, V1 LXVD2X (CTX)(R_x030), VS38 // v6 = vs38 VSLDOI $8, V2, V2, V3 VSLDOI $8, V4, V4, V5 VSLDOI $8, V6, V6, V7 loop: MOVD TBL_STRT, TBL LVX (TBL)(R_x000), KI LXVD2X (INP)(R0), VS40 // load v8 (=vs40) in advance ADD $16, INP // Copy V0-V7 to VS24-VS31 XXLOR V0, V0, VS24 XXLOR V1, V1, VS25 XXLOR V2, V2, VS26 XXLOR V3, V3, VS27 XXLOR V4, V4, VS28 XXLOR V5, V5, VS29 XXLOR V6, V6, VS30 XXLOR V7, V7, VS31 VADDUDM KI, V7, V7 // h+K[i] LVX (TBL)(R_x010),...