loop:
	QR(X0, X4, X8, X12, X15)
	MOVOU X4, (4*16)(BX) // save X4
	QR(X1, X5, X9, X13, X15)
	MOVOU (15*16)(BX), X15 // reload X15; temp now X4
	QR(X2, X6, X10, X14, X4)
	QR(X3, X7, X11, X15, X4)

	QR(X0, X5, X10, X15, X4)
	MOVOU X15, (15*16)(BX) // save X15
	QR(X1, X6, X11, X12, X4)
	MOVOU (4*16)(BX), X4  // reload X4; temp now X15
	QR(X2, X7, X8, X13, X15)
	QR(X3, X4, X9, X14, X15)

	DECL DX
	JNZ loop

		x1 := byteorder.LeUint32(q[4*0x1:])
		x2 := byteorder.LeUint32(q[4*0x2:])
		x3 := byteorder.LeUint32(q[4*0x3:])
		x4 := byteorder.LeUint32(q[4*0x4:])
		x5 := byteorder.LeUint32(q[4*0x5:])
		x6 := byteorder.LeUint32(q[4*0x6:])
		x7 := byteorder.LeUint32(q[4*0x7:])
		x8 := byteorder.LeUint32(q[4*0x8:])
		x9 := byteorder.LeUint32(q[4*0x9:])
		xa := byteorder.LeUint32(q[4*0xa:])
		xb := byteorder.LeUint32(q[4*0xb:])

hugeloop:
	CMPQ	BX, $64
	JB	bigloop
	MOVOU	(SI), X0
	MOVOU	(DI), X1
	MOVOU	16(SI), X2
	MOVOU	16(DI), X3
	MOVOU	32(SI), X4
	MOVOU	32(DI), X5
	MOVOU	48(SI), X6
	MOVOU	48(DI), X7
	PCMPEQB	X1, X0
	PCMPEQB	X3, X2
	PCMPEQB	X5, X4
	PCMPEQB	X7, X6
	PAND	X2, X0
	PAND	X6, X4
	PAND	X4, X0
	PMOVMSKB X0, DX
	ADDQ	$64, SI
	ADDQ	$64, DI
	SUBQ	$64, BX
	CMPL	DX, $0xffff
	JEQ	hugeloop
	XORQ	AX, AX	// return 0
	RET

	// If less than 8 bytes, do single byte copies.
	MOV	$8, X9
	BLT	X12, X9, f_loop4_check

	// Check alignment - if alignment differs we have to do one byte at a time.
	AND	$7, X10, X5
	AND	$7, X11, X6
	BNE	X5, X6, f_loop8_unaligned_check
	BEQZ	X5, f_loop_check

	// Move one byte at a time until we reach 8 byte alignment.
	SUB	X5, X9, X5
	SUB	X5, X12, X12
f_align:
	SUB	$1, X5
	MOVB	0(X11), X14
	MOVB	X14, 0(X10)

	MOVOU z1in(16*0), X13
	MOVOU z1in(16*1), X14

	PAND X15, X0
	PAND X15, X1
	PAND X15, X2
	PAND X15, X3
	PAND X15, X4
	PAND X15, X5

	PAND X6, X9
	PAND X6, X10
	PAND X6, X11
	PAND X6, X12
	PAND X6, X13
	PAND X6, X14

	PXOR X9, X0
	PXOR X10, X1
	PXOR X11, X2
	PXOR X12, X3
	PXOR X13, X4
	PXOR X14, X5
	// Similarly if zero == 0
	PCMPEQL X9, X9

	MOVOU	16(AX), X5
	MOVOU	-32(AX)(BX*1), X6
	MOVOU	-16(AX)(BX*1), X7

	PXOR	X0, X4
	PXOR	X1, X5
	PXOR	X2, X6
	PXOR	X3, X7

	AESENC	X4, X4
	AESENC	X5, X5
	AESENC	X6, X6
	AESENC	X7, X7

	AESENC	X4, X4
	AESENC	X5, X5
	AESENC	X6, X6
	AESENC	X7, X7

	AESENC	X4, X4
	AESENC	X5, X5
	AESENC	X6, X6
	AESENC	X7, X7

	PXOR	X6, X4

				{1, 65280}, // X0 X1 X2 X3 X4 X5 X6 X7
			},
			outputs: []outputInfo{
				{0, 65280}, // X0 X1 X2 X3 X4 X5 X6 X7
			},
		},
	},
	{
		name:         "SUBSD",
		argLen:       2,
		resultInArg0: true,
		asm:          x86.ASUBSD,
		reg: regInfo{
			inputs: []inputInfo{
				{0, 65280}, // X0 X1 X2 X3 X4 X5 X6 X7
				{1, 65280}, // X0 X1 X2 X3 X4 X5 X6 X7
			},

// Code generated by mkpreempt.go; DO NOT EDIT.

#include "go_asm.h"
#include "textflag.h"

TEXT ·asyncPreempt(SB),NOSPLIT|NOFRAME,$0-0
	MOV X1, -464(X2)
	SUB $464, X2
	MOV X5, 8(X2)
	MOV X6, 16(X2)
	MOV X7, 24(X2)
	MOV X8, 32(X2)
	MOV X9, 40(X2)
	MOV X10, 48(X2)
	MOV X11, 56(X2)
	MOV X12, 64(X2)
	MOV X13, 72(X2)
	MOV X14, 80(X2)
	MOV X15, 88(X2)
	MOV X16, 96(X2)
	MOV X17, 104(X2)
	MOV X18, 112(X2)

	SHUFPS $0x10, X0, X4
	PXOR X4, X0
	SHUFPS $0x8c, X0, X4
	PXOR X4, X0
	PXOR X1, X0

	MOVAPS X2, X5
	MOVAPS X2, X6
	PSLLDQ $0x4, X5
	PSHUFD $0xff, X0, X3
	PXOR X3, X2
	PXOR X5, X2

	MOVAPS X0, X1
	SHUFPS $0x44, X0, X6
	MOVUPS X6, (BX)
	SHUFPS $0x4e, X2, X1
	MOVUPS X1, 16(BX)
	ADDQ $32, BX
	RET

TEXT _expand_key_192b<>(SB),NOSPLIT,$0
	PSHUFD $0x55, X1, X1

	MOVOU	16(AX), X5
	MOVOU	-32(AX)(CX*1), X6
	MOVOU	-16(AX)(CX*1), X7

	PXOR	X0, X4
	PXOR	X1, X5
	PXOR	X2, X6
	PXOR	X3, X7

	AESENC	X4, X4
	AESENC	X5, X5
	AESENC	X6, X6
	AESENC	X7, X7

	AESENC	X4, X4
	AESENC	X5, X5
	AESENC	X6, X6
	AESENC	X7, X7

	AESENC	X4, X4
	AESENC	X5, X5
	AESENC	X6, X6
	AESENC	X7, X7

	PXOR	X6, X4