SHA512ROUND(V1, V4, V3, V0, V2, V24, V28, V16, V17, V15, V12, V13)

	SHA512ROUND(V0, V1, V2, V3, V4, V25, V29, V17, V18, V16, V13, V14)
	SHA512ROUND(V3, V0, V4, V2, V1, V26, V30, V18, V19, V17, V14, V15)
	SHA512ROUND(V2, V3, V1, V4, V0, V27, V31, V19, V12, V18, V15, V16)
	SHA512ROUND(V4, V2, V0, V1, V3, V28, V24, V12, V13, V19, V16, V17)
	SHA512ROUND(V1, V4, V3, V0, V2, V29, V25, V13, V14, V12, V17, V18)

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

	VXOR V4, V8, V4
	VXOR V5, V9, V5
	VXOR V6, V10, V6
	VXOR V7, V11, V7

	VRLW V4, V28, V4
	VRLW V5, V28, V5
	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

	VZIP2	V26.D2, V24.D2, V6.D2
	VZIP1	V27.D2, V25.D2, V10.D2
	VZIP2	V27.D2, V25.D2, V14.D2
	VLD1.P	64(R2), [V24.B16, V25.B16, V26.B16, V27.B16]
	VZIP1	V30.D2, V28.D2, V3.D2
	VZIP2	V30.D2, V28.D2, V7.D2
	VZIP1	V31.D2, V29.D2, V11.D2
	VZIP2	V31.D2, V29.D2, V15.D2
	VLD1.P	64(R2), [V28.B16, V29.B16, V30.B16, V31.B16]
	VEOR	V0.B16, V16.B16, V16.B16
	VEOR	V1.B16, V17.B16, V17.B16
	VEOR	V2.B16, V18.B16, V18.B16
	VEOR	V3.B16, V19.B16, V19.B16

    T26 v26, T27 v27, T28 v28) {
  return internal::ValueArray28<T1, T2, T3, T4, T5, T6, T7, T8, T9, T10, T11,
      T12, T13, T14, T15, T16, T17, T18, T19, T20, T21, T22, T23, T24, T25,
      T26, T27, T28>(v1, v2, v3, v4, v5, v6, v7, v8, v9, v10, v11, v12, v13,
      v14, v15, v16, v17, v18, v19, v20, v21, v22, v23, v24, v25, v26, v27,
      v28);
}

    T26 v26, T27 v27, T28 v28) {
  return internal::ValueArray28<T1, T2, T3, T4, T5, T6, T7, T8, T9, T10, T11,
      T12, T13, T14, T15, T16, T17, T18, T19, T20, T21, T22, T23, T24, T25,
      T26, T27, T28>(v1, v2, v3, v4, v5, v6, v7, v8, v9, v10, v11, v12, v13,
      v14, v15, v16, v17, v18, v19, v20, v21, v22, v23, v24, v25, v26, v27,
      v28);
}

int foo_gpr0() {
        asm volatile("":::"r30","r29","r28");
}
int foo_gpr1() {
        asm volatile("":::"fr31", "fr30","fr29","fr28","r30","r29","r28");
}
int foo_vr() {
        asm volatile("":::"v31","v30","v29","v28");
}
*/
import "C"

import "fmt"

func main() {
	C.foo_fpr()
	C.foo_gpr0()
	C.foo_gpr1()
	C.foo_vr()
	fmt.Println("success")

      T26 v26, T27 v27, T28 v28) : v1_(v1), v2_(v2), v3_(v3), v4_(v4), v5_(v5),
      v6_(v6), v7_(v7), v8_(v8), v9_(v9), v10_(v10), v11_(v11), v12_(v12),
      v13_(v13), v14_(v14), v15_(v15), v16_(v16), v17_(v17), v18_(v18),
      v19_(v19), v20_(v20), v21_(v21), v22_(v22), v23_(v23), v24_(v24),
      v25_(v25), v26_(v26), v27_(v27), v28_(v28) {}

  template <typename T>

      T26 v26, T27 v27, T28 v28) : v1_(v1), v2_(v2), v3_(v3), v4_(v4), v5_(v5),
      v6_(v6), v7_(v7), v8_(v8), v9_(v9), v10_(v10), v11_(v11), v12_(v12),
      v13_(v13), v14_(v14), v15_(v15), v16_(v16), v17_(v17), v18_(v18),
      v19_(v19), v20_(v20), v21_(v21), v22_(v22), v23_(v23), v24_(v24),
      v25_(v25), v26_(v26), v27_(v27), v28_(v28) {}

  template <typename T>

	VPDI $0x4, V13, V13, V13 // flip the doublewords to big-endian order
	VPDI $0x4, V14, V14, V14 // flip the doublewords to big-endian order

	VACCCQ V5, V13, V28, V29
	VACQ   V5, V13, V28, V21
	VACCCQ V6, V14, V29, V30
	VACQ   V6, V14, V29, V22

	VPDI $0x4, V7, V7, V7    // flip the doublewords to big-endian order
	VPDI $0x4, V8, V8, V8    // flip the doublewords to big-endian order

	VLD1.P	64(R2), [V16.S4, V17.S4, V18.S4, V19.S4]
	VLD1.P	64(R2), [V20.S4, V21.S4, V22.S4, V23.S4]
	VLD1.P	64(R2), [V24.S4, V25.S4, V26.S4, V27.S4]
	VLD1	(R2), [V28.S4, V29.S4, V30.S4, V31.S4]     //load 64*4bytes K constant(K0-K63)

blockloop:

	VLD1.P	16(R1), [V4.B16]                            // load 16bytes message