MOVOU	96(SI), X6
	MOVOU	112(SI), X7
	MOVOU	128(SI), X8
	MOVOU	144(SI), X9
	MOVOU	160(SI), X10
	MOVOU	176(SI), X11
	MOVOU	192(SI), X12
	MOVOU	208(SI), X13
	MOVOU	224(SI), X14
	MOVOU	240(SI), X15
	MOVOU	X0, (DI)
	MOVOU	X1, 16(DI)
	MOVOU	X2, 32(DI)
	MOVOU	X3, 48(DI)
	MOVOU	X4, 64(DI)
	MOVOU	X5, 80(DI)
	MOVOU	X6, 96(DI)
	MOVOU	X7, 112(DI)
	MOVOU	X8, 128(DI)
	MOVOU	X9, 144(DI)
	MOVOU	X10, 160(DI)

	#endif
	VZEROUPPER
	#endif
	MOVUPS X0, 112(SP)
	MOVUPS X1, 128(SP)
	MOVUPS X2, 144(SP)
	MOVUPS X3, 160(SP)
	MOVUPS X4, 176(SP)
	MOVUPS X5, 192(SP)
	MOVUPS X6, 208(SP)
	MOVUPS X7, 224(SP)
	MOVUPS X8, 240(SP)
	MOVUPS X9, 256(SP)
	MOVUPS X10, 272(SP)
	MOVUPS X11, 288(SP)
	MOVUPS X12, 304(SP)
	MOVUPS X13, 320(SP)
	MOVUPS X14, 336(SP)
	MOVUPS X15, 352(SP)
	CALL ·asyncPreempt2(SB)
	MOVUPS 352(SP), X15

// Switches SP to g0 stack and calls (X14). Arguments already set.
TEXT	asancall<>(SB), NOSPLIT, $0-0
	MOV	X2, X8		// callee-saved
	BEQZ	g, g0stack	// no g, still on a system stack
	MOV	g_m(g), X21
	MOV	m_g0(X21), X21
	BEQ	X21, g, g0stack

	MOV	(g_sched+gobuf_sp)(X21), X2

g0stack:
	JALR	RA, X14
	MOV	X8, X2

// an extra extension is usually needed due to
// no signed byte load.

func set16(x8 int8, u8 *uint8, y8 int8, z8 uint8) {
	// Truncate not needed, load does sign/zero extend

	// ppc64x:-"MOVBZ\tR\\d+,\\sR\\d+"
	val16[0] = uint16(*u8)

	// AND not needed due to size
	// ppc64x:-"ANDCC"
	sval16[1] = 255 & int16(x8+y8)

	// ppc64x:-"ANDCC"
	val16[1] = 255 & uint16(*u8+z8)

}

	VMOVDQU32 X8, K3, X19                              // 62317e0b7fc3
	VMOVDQU32 X8, K3, (R14)                            // 62517e0b7f06
	VMOVDQU32 X8, K3, -7(DI)(R8*8)                     // 62317e0b7f84c7f9ffffff
	VMOVDQU32 X26, K4, X8                              // 62417e0c7fd0
	VMOVDQU32 99(R15)(R15*4), K4, X8                   // 62117e0c6f84bf63000000

// Copyright 2022 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.

package abi

const (
	// See abi_generic.go.

	// X8 - X23
	IntArgRegs = 16

	// F8 - F23.
	FloatArgRegs = 16

	EffectiveFloatRegSize = 8

	// scramble 3 times
	AESENC	X8, X8
	AESENC	X9, X9
	AESENC	X10, X10
	AESENC	X11, X11
	AESENC	X12, X12
	AESENC	X13, X13
	AESENC	X14, X14
	AESENC	X15, X15

	AESENC	X8, X8
	AESENC	X9, X9
	AESENC	X10, X10
	AESENC	X11, X11
	AESENC	X12, X12
	AESENC	X13, X13
	AESENC	X14, X14
	AESENC	X15, X15

	AESENC	X8, X8
	AESENC	X9, X9
	AESENC	X10, X10

	MOVQ	BP, (2*8)(SP)	\
	MOVQ	BX, (3*8)(SP)	\
	MOVQ	R12, (4*8)(SP)	\
	MOVQ	R13, (5*8)(SP)	\
	MOVQ	R14, (6*8)(SP)	\
	MOVQ	R15, (7*8)(SP)	\
	MOVUPS	X6, (8*8)(SP)	\
	MOVUPS	X7, (10*8)(SP)	\
	MOVUPS	X8, (12*8)(SP)	\
	MOVUPS	X9, (14*8)(SP)	\
	MOVUPS	X10, (16*8)(SP)	\
	MOVUPS	X11, (18*8)(SP)	\
	MOVUPS	X12, (20*8)(SP)	\
	MOVUPS	X13, (22*8)(SP)	\
	MOVUPS	X14, (24*8)(SP)	\
	MOVUPS	X15, (26*8)(SP)

//line x8.go:4
package main

// When building with -buildmode=c-shared, this symbol is called when the shared
// library is loaded.
TEXT _rt0_riscv64_freebsd_lib(SB),NOSPLIT,$224
	// Preserve callee-save registers, along with X1 (LR).
	MOV	X1, (8*3)(X2)
	MOV	X8, (8*4)(X2)
	MOV	X9, (8*5)(X2)
	MOV	X18, (8*6)(X2)
	MOV	X19, (8*7)(X2)
	MOV	X20, (8*8)(X2)
	MOV	X21, (8*9)(X2)
	MOV	X22, (8*10)(X2)
	MOV	X23, (8*11)(X2)
	MOV	X24, (8*12)(X2)
	MOV	X25, (8*13)(X2)