// and corresponds to first matching byte (bit 6 is 0)
	PCMPESTRI $0x0c, (DI), X1
	// CX == 16 means no match,
	// CX > R9 means partial match at the end of the string,
	// otherwise sep is at offset CX from X1 start
	CMPQ CX, R9
	JBE sse42_success
	ADDQ R9, DI
	CMPQ DI, SI
	JB loop_sse42
	PCMPESTRI $0x0c, -1(SI), X1
	CMPQ CX, R9
	JA fail
	LEAQ -1(SI), DI
sse42_success:

	"PADDW",
	"PALIGNR",
	"PAND",
	"PANDN",
	"PAUSE",
	"PAVGB",
	"PAVGW",
	"PBLENDVB",
	"PBLENDW",
	"PCLMULQDQ",
	"PCMPEQB",
	"PCMPEQL",
	"PCMPEQQ",
	"PCMPEQW",
	"PCMPESTRI",
	"PCMPESTRM",
	"PCMPGTB",
	"PCMPGTL",
	"PCMPGTQ",
	"PCMPGTW",
	"PCMPISTRI",
	"PCMPISTRM",
	"PDEPL",
	"PDEPQ",
	"PEXTL",
	"PEXTQ",
	"PEXTRB",
	"PEXTRD",
	"PEXTRQ",

	PCMPESTRI $7, (BX), X2                  // 660f3a611307
	PCMPESTRI $7, (R11), X2                 // 66410f3a611307
	PCMPESTRI $7, X2, X2                    // 660f3a61d207
	PCMPESTRI $7, X11, X2                   // 66410f3a61d307
	PCMPESTRI $7, (BX), X11                 // 66440f3a611b07
	PCMPESTRI $7, (R11), X11                // 66450f3a611b07
	PCMPESTRI $7, X2, X11                   // 66440f3a61da07

	PBLENDW:         "PBLENDW",
	PCLMULQDQ:       "PCLMULQDQ",
	PCMPEQB:         "PCMPEQB",
	PCMPEQD:         "PCMPEQD",
	PCMPEQQ:         "PCMPEQQ",
	PCMPEQW:         "PCMPEQW",
	PCMPESTRI:       "PCMPESTRI",
	PCMPESTRM:       "PCMPESTRM",
	PCMPGTB:         "PCMPGTB",
	PCMPGTD:         "PCMPGTD",
	PCMPGTQ:         "PCMPGTQ",
	PCMPGTW:         "PCMPGTW",
	PCMPISTRI:       "PCMPISTRI",