VXOR	V19,V15,V19
v14:	VXOR	V20,V14,V20
v13:	VXOR	V19,V13,V19
v12:	VXOR	V20,V12,V20
v11:	VXOR	V19,V11,V19
v10:	VXOR	V20,V10,V20
v9:	VXOR	V19,V9,V19
v8:	VXOR	V20,V8,V20
v7:	VXOR	V19,V7,V19
v6:	VXOR	V20,V6,V20
v5:	VXOR	V19,V5,V19
v4:	VXOR	V20,V4,V20
v3:	VXOR	V19,V3,V19
v2:	VXOR	V20,V2,V20
v1:	VXOR	V19,V1,V19
v0:	VXOR	V20,V0,V20

	VXOR	V19,V20,V0

	BR	barrett_reduction

            validateStringNoExpression("groupId", problems, Severity.WARNING, Version.V20, m.getGroupId(), m);
            if (parent == null) {
                validateStringNotEmpty("groupId", problems, Severity.FATAL, Version.V20, m.getGroupId(), m);
            }

            validateStringNoExpression("artifactId", problems, Severity.WARNING, Version.V20, m.getArtifactId(), m);

            validateStringNoExpression("groupId", problems, Severity.WARNING, Version.V20, m.getGroupId(), m);
            if (parent == null) {
                validateStringNotEmpty("groupId", problems, Severity.FATAL, Version.V20, m.getGroupId(), m);
            }

            validateStringNoExpression("artifactId", problems, Severity.WARNING, Version.V20, m.getArtifactId(), m);

    T14 v14, T15 v15, T16 v16, T17 v17, T18 v18, T19 v19, T20 v20) {
  return internal::ValueArray20<T1, T2, T3, T4, T5, T6, T7, T8, T9, T10, T11,
      T12, T13, T14, T15, T16, T17, T18, T19, T20>(v1, v2, v3, v4, v5, v6, v7,
      v8, v9, v10, v11, v12, v13, v14, v15, v16, v17, v18, v19, v20);
}

template <typename T1, typename T2, typename T3, typename T4, typename T5,

	SHA256ROUND0(V6, V7, V0, V1, V2, V3, V4, V5, V18, R_x0c0)
	VSLDOI	$4, V18, V18, V19
	LXVD2X	(INP)(R_x030), V20 // load v20 in advance
	SHA256ROUND0(V5, V6, V7, V0, V1, V2, V3, V4, V19, R_x0d0)
	VPERMLE(V20, V20, LEMASK, V20)
	SHA256ROUND0(V4, V5, V6, V7, V0, V1, V2, V3, V20, R_x0e0)
	VSLDOI	$4, V20, V20, V21
	SHA256ROUND0(V3, V4, V5, V6, V7, V0, V1, V2, V21, R_x0f0)
	VSLDOI	$4, V21, V21, V22

	VRLW V6, V28, V6
	VRLW V7, V28, V7

	VADDUWM V0, V4, V0
	VADDUWM V1, V5, V1
	VADDUWM V2, V6, V2
	VADDUWM V3, V7, V3

	VPERMXOR V12, V0, V20, V12
	VPERMXOR V13, V1, V20, V13
	VPERMXOR V14, V2, V20, V14
	VPERMXOR V15, V3, V20, V15

	VADDUWM V8, V12, V8
	VADDUWM V9, V13, V9
	VADDUWM V10, V14, V10
	VADDUWM V11, V15, V11

	VXOR V4, V8, V4
	VXOR V5, V9, V5
	VXOR V6, V10, V6

	LXVD2X	(INP)(R_x050), VS52	// load v20 (=vs52) in advance
	VSLDOI	$8, V18, V18, V19
	SHA512ROUND0(V5, V6, V7, V0, V1, V2, V3, V4, V19, R_x0d0)
	VPERMLE(V20,V20,LEMASK,V20)
	SHA512ROUND0(V4, V5, V6, V7, V0, V1, V2, V3, V20, R_x0e0)
	LXVD2X	(INP)(R_x060), VS54	// load v22 (=vs54) in advance
	VSLDOI	$8, V20, V20, V21
	SHA512ROUND0(V3, V4, V5, V6, V7, V0, V1, V2, V21, R_x0f0)

	VADD	V18.S4, V7.S4, V16.S4
	HASHUPDATEPARITY

	SUB	$64, R3, R3                                  // message length - 64bytes, then compare with 64bytes
	VADD	V2.S4, V0.S4, V0.S4
	VADD	V1.S4, V20.S4, V20.S4
	CBNZ	R3, blockloop

sha1ret:

	VST1.P	[V0.S4], 16(R0)                               // store hash value H(dcba)
	FMOVS	F20, (R0)                                     // store hash value H(e)

	VREV64	V16.B16, V16.B16
	VREV64	V17.B16, V17.B16
	VREV64	V18.B16, V18.B16
	VREV64	V19.B16, V19.B16

	MOVD	R3, R4
	// load first 4 round consts in V20-V23
	VLD1.P	64(R4), [V20.D2, V21.D2, V22.D2, V23.D2]

	SHA512ROUND(V0, V1, V2, V3, V4, V20, V24, V12, V13, V19, V16, V17)
	SHA512ROUND(V3, V0, V4, V2, V1, V21, V25, V13, V14, V12, V17, V18)
	SHA512ROUND(V2, V3, V1, V4, V0, V22, V26, V14, V15, V13, V18, V19)

// the expanded keys are loaded, it is not needed.
//
// R12,R14-R21 are scratch registers.
// For keyp of 10, V6, V11-V20 hold the expanded key.
// For keyp of 12, V6, V9-V20 hold the expanded key.
// For keyp of 14, V6, V7-V20 hold the expanded key.
#define LOAD_KEY(Rkeyp) \
	MOVD	$16, R12 \
	MOVD	$32, R14 \
	MOVD	$48, R15 \
	MOVD	$64, R16 \
	MOVD	$80, R17 \
	MOVD	$96, R18 \