ROUND2(R22,R3,R4,R5,M01,0xf61e2562, 5);
	ROUND2(R5,R22,R3,R4,M06,0xc040b340, 9);
	ROUND2(R4,R5,R22,R3,M11,0x265e5a51,14);
	ROUND2(R3,R4,R5,R22,M00,0xe9b6c7aa,20);
	ROUND2(R22,R3,R4,R5,M05,0xd62f105d, 5);
	ROUND2(R5,R22,R3,R4,M10, 0x2441453, 9);
	ROUND2(R4,R5,R22,R3,M15,0xd8a1e681,14);
	ROUND2(R3,R4,R5,R22,M04,0xe7d3fbc8,20);
	ROUND2(R22,R3,R4,R5,M09,0x21e1cde6, 5);
	ROUND2(R5,R22,R3,R4,M14,0xc33707d6, 9);

				{1, 670826495}, // R0 R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R19 R20 R21 R22 R23 R24 R25 R26 R30
			},
			outputs: []outputInfo{
				{1, 0},
				{0, 670826495}, // R0 R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R19 R20 R21 R22 R23 R24 R25 R26 R30
			},
		},
	},
	{

	// We won't get a nested signal.
	MOVBZ	runtime·iscgo(SB), R22
	CMP	R22, $0
	BNE	nosaveg
	MOVD	m_gsignal(R21), R22	// g.m.gsignal
	CMP	R22, $0
	BEQ	nosaveg

	CMP	g, R22
	BEQ	nosaveg
	MOVD	(g_stack+stack_lo)(R22), R22 // g.m.gsignal.stack.lo
	MOVD	g, (R22)

	BL	(CTR)	// Call from VDSO

	MOVD	$0, (R22)	// clear g slot, R22 is unchanged by C code

	JMP	finish

nosaveg:

	MOVD    16(R8), R21       // R21 = x[i+1]
	MOVD    24(R8), R22       // R22 = x[i+2]
	MOVDU   32(R8), R23       // R23 = x[i+3]
	MULLD   R9, R20, R24      // R24 = z0[i]
	MULHDU  R9, R20, R20      // R20 = z1[i]
	ADDC    R4, R24           // R24 = z0[i] + c
	MULLD   R9, R21, R25
	MULHDU  R9, R21, R21
	ADDE    R20, R25
	MULLD   R9, R22, R26
	MULHDU  R9, R22, R22
	MULLD   R9, R23, R27
	MULHDU  R9, R23, R23

	MOVD	R21, 216(R1)
	MOVD	R22, 224(R1)
	MOVD	R23, 232(R1)
	MOVD	R24, 240(R1)
	MOVD	R25, 248(R1)
	MOVD	R26, 256(R1)
	MOVD	R27, 264(R1)
	MOVD	R28, 272(R1)
	MOVD	R29, 280(R1)
	MOVD	g, 288(R1)
	MOVD	LR, R31
	MOVD	R31, 32(R1)
	CALL	runtime·debugCallCheck(SB)
	MOVD	40(R1), R22
	XOR	R0, R0
	CMP	R22, $0
	BEQ	good
	MOVD	48(R1), R22
	MOVD	$8, R20
	TW	$31, R0, R0

	SLD	$2,R18,R18	// p[6]*4
	MOVWZ	(R10)(R17),R21	// tab[0][p[7]]
	ADD	$1024,R10,R10	// tab[1]
	RLDICL	$56,R9,$56,R19	// p[5]
	SLD	$2,R19,R19	// p[5]*4:1
	MOVWZ	(R10)(R18),R22	// tab[1][p[6]]
	ADD	$1024,R10,R10	// tab[2]
	XOR	R21,R22,R21	// xor done R22
	CLRLSLDI $56,R9,$2,R20
	MOVWZ	(R10)(R19),R23	// tab[2][p[5]]
	ADD	$1024,R10,R10	// &tab[3]
	XOR	R21,R23,R21	// xor done R23
	MOVWZ	(R10)(R20),R24	// tab[3][p[4]]

				// if(g->panic != nil && g->panic->argp == FP) g->panic->argp = bottom-of-frame
				//
				//	MOVD g_panic(g), R22
				//	CMP R22, $0
				//	BEQ end
				//	MOVD panic_argp(R22), R23
				//	ADD $(autosize+8), R1, R24
				//	CMP R23, R24
				//	BNE end
				//	ADD $8, R1, R25
				//	MOVD R25, panic_argp(R22)
				// end:
				//	NOP
				//
				// The NOP is needed to give the jumps somewhere to land.

	MOVW	R13, 52(R29)
	MOVW	R14, 56(R29)
	MOVW	R15, 60(R29)
	MOVW	R16, 64(R29)
	MOVW	R17, 68(R29)
	MOVW	R18, 72(R29)
	MOVW	R19, 76(R29)
	MOVW	R20, 80(R29)
	// R21 already saved
	// R22 already saved.
	MOVW	R22, 84(R29)
	// R23 is tmp register.
	MOVW	R24, 88(R29)
	MOVW	R25, 92(R29)
	// R26 is reserved by kernel.
	// R27 is reserved by kernel.
	MOVW	R28, 96(R29)
	// R29 is SP.
	// R30 is g.

	STP	(R10, R11), 9*8(RSP)
	STP	(R12, R13), 11*8(RSP)
	STP	(R14, R15), 13*8(RSP)
	// R16, R17 may be clobbered by linker trampoline
	// R18 is unused.
	STP	(R19, R20), 15*8(RSP)
	STP	(R21, R22), 17*8(RSP)
	STP	(R23, R24), 19*8(RSP)
	STP	(R25, R26), 21*8(RSP)
	// R27 is temp register.
	// R28 is g.
	// R29 is frame pointer (unused).
	// R30 is LR, which was saved by the prologue.
	// R31 is SP.

TEXT runtime·return0(SB), NOSPLIT, $0
	MOVW	$0, R19
	RET

// Called from cgo wrappers, this function returns g->m->curg.stack.hi.
// Must obey the gcc calling convention.
TEXT _cgo_topofstack(SB),NOSPLIT,$16
	// g (R22) and REGTMP (R30)  might be clobbered by load_g. They
	// are callee-save in the gcc calling convention, so save them.
	MOVV	R30, savedREGTMP-16(SP)
	MOVV	g, savedG-8(SP)

	JAL	runtime·load_g(SB)
	MOVV	g_m(g), R19