ADCXQ BX, R8
	ADOXQ 32(AX), R8
	MOVQ  R8, 32(AX)

	// Iteration 5
	MULXQ 40(CX), R8, BX
	ADCXQ DI, R8
	ADOXQ 40(AX), R8
	MOVQ  R8, 40(AX)

	// Iteration 6
	MULXQ 48(CX), R8, DI
	ADCXQ BX, R8
	ADOXQ 48(AX), R8
	MOVQ  R8, 48(AX)

	// Iteration 7
	MULXQ 56(CX), R8, BX
	ADCXQ DI, R8
	ADOXQ 56(AX), R8
	MOVQ  R8, 56(AX)

	// Iteration 8
	MULXQ 64(CX), R8, DI

DATA zosLibVec<>(SB)/8, $0
GLOBL zosLibVec<>(SB), NOPTR, $8

TEXT ·initZosLibVec(SB), NOSPLIT|NOFRAME, $0-0
	MOVW PSALAA, R8
	MOVD LCA64(R8), R8
	MOVD CAA(R8), R8
	MOVD EDCHPXV(R8), R8
	MOVD R8, zosLibVec<>(SB)
	RET

TEXT ·GetZosLibVec(SB), NOSPLIT|NOFRAME, $0-0
	MOVD zosLibVec<>(SB), R8
	MOVD R8, ret+0(FP)
	RET

TEXT ·clearErrno(SB), NOSPLIT, $0-0
	BL   addrerrno<>(SB)
	MOVD $0, 0(R3)
	RET

largeLoop:  // Copying 256 bytes at a time.
	MVC    $256, 0(R6), 0(R8)
	MOVD   $256(R6), R6
	MOVD   $256(R8), R8
	MOVD   $-32(R5), R5
	CMPBGE R5, $32, largeLoop
	BR     mediumLoop

mediumLoopBody:  // Copying 32 bytes at a time
	MVC    $32, 0(R6), 0(R8)
	MOVD   $32(R6), R6
	MOVD   $32(R8), R8
	MOVD   $-4(R5), R5
	CMPBGE R5, $4, mediumLoopBody
	BR     smallLoop

returnC:

	MOVW	R0, R8

	MOVW	(0*4)(R8), R7 // fn
	MOVW	(1*4)(R8), R0 // a1
	MOVW	(2*4)(R8), R1 // a2
	MOVW	(3*4)(R8), R2 // a3
	MOVW	(4*4)(R8), R3 // a4
	MOVW	(5*4)(R8), R4 // a5
	MOVW	R4, 0(R13)
	MOVW	(6*4)(R8), R5 // a6
	MOVW	R5, 4(R13)
	MOVW	(7*4)(R8), R6 // a7
	MOVW	R6, 8(R13)
	MOVW	(8*4)(R8), R4 // a8
	MOVW	R4, 12(R13)
	MOVW	(9*4)(R8), R5 // a9
	MOVW	R5, 16(R13)

	SHA256ROUND0(3, 0xe9b5dba5, R13, R14, R15, R8, R9, R10, R11, R12)
	SHA256ROUND0(4, 0x3956c25b, R12, R13, R14, R15, R8, R9, R10, R11)
	SHA256ROUND0(5, 0x59f111f1, R11, R12, R13, R14, R15, R8, R9, R10)
	SHA256ROUND0(6, 0x923f82a4, R10, R11, R12, R13, R14, R15, R8, R9)
	SHA256ROUND0(7, 0xab1c5ed5, R9, R10, R11, R12, R13, R14, R15, R8)
	SHA256ROUND0(8, 0xd807aa98, R8, R9, R10, R11, R12, R13, R14, R15)

	VPMADD52HUQ -7(DI)(R8*4), X11, K1, X18             // 62a2a509b59487f9ffffff
	VPMADD52HUQ X7, X31, K1, X18                       // 62e28501b5d7
	VPMADD52HUQ X0, X31, K1, X18                       // 62e28501b5d0
	VPMADD52HUQ 17(SP)(BP*2), X31, K1, X18             // 62e28501b5946c11000000
	VPMADD52HUQ -7(DI)(R8*4), X31, K1, X18             // 62a28501b59487f9ffffff

	LSR	R4, R6, R5	// return value
	LSL	R3, R6, R8	// x[i] << s
	SUB	$1, R1
one:	TBZ	$0, R1, two
	MOVD.W	-8(R2), R6
	LSR	R4, R6, R7
	ORR	R8, R7
	LSL	R3, R6, R8
	SUB	$1, R1
	MOVD.W	R7, -8(R0)
two:
	TBZ	$1, R1, loop
	LDP.W	-16(R2), (R6, R7)
	LSR	R4, R7, R10
	ORR	R8, R10
	LSL	R3, R7
	LSR	R4, R6, R9
	ORR	R7, R9
	LSL	R3, R6, R8
	SUB	$2, R1
	STP.W	(R9, R10), -16(R0)
loop:

	VGF2P8AFFINEQB $82, -7(DI)(R8*4), X21, K1, X15     // 6233d501cebc87f9ffffff52
	VGF2P8AFFINEQB $82, X22, X0, K1, X15               // 6233fd09cefe52
	VGF2P8AFFINEQB $82, X7, X0, K1, X15                // 6273fd09ceff52
	VGF2P8AFFINEQB $82, X19, X0, K1, X15               // 6233fd09cefb52
	VGF2P8AFFINEQB $82, 17(SP)(BP*2), X0, K1, X15      // 6273fd09cebc6c1100000052
	VGF2P8AFFINEQB $82, -7(DI)(R8*4), X0, K1, X15      // 6233fd09cebc87f9ffffff52

	SHA512ROUND0(14, 0x9bdc06a725c71235, R10, R11, R12, R13, R14, R15, R8, R9)
	SHA512ROUND0(15, 0xc19bf174cf692694, R9, R10, R11, R12, R13, R14, R15, R8)

	SHA512ROUND1(16, 0xe49b69c19ef14ad2, R8, R9, R10, R11, R12, R13, R14, R15)
	SHA512ROUND1(17, 0xefbe4786384f25e3, R15, R8, R9, R10, R11, R12, R13, R14)
	SHA512ROUND1(18, 0x0fc19dc68b8cd5b5, R14, R15, R8, R9, R10, R11, R12, R13)

	BL runtime·mstart(SB)

	MOVW $2, R8  // crash (not reached)
	MOVW R8, (R8)
	RET

// Exit the entire program (like C exit)
TEXT runtime·exit(SB),NOSPLIT|NOFRAME,$0
	MOVW code+0(FP), R0	// arg 1 exit status
	MOVW $SYS_exit, R7
	SWI $0
	MOVW.CS $0, R8 // crash on syscall failure
	MOVW.CS R8, (R8)
	RET

// func exitThread(wait *atomic.Uint32)