issue with mbx_sm4_xts_encrypt/decrypt_mb16

[Jingkai]

Attempted to use the functions mbx_sm4_xts_encrypt/decrypt_mb16 for data encryption and decryption, but found inconsistencies in the data before encryption and after decryption.

TRANSPOSE_OUT_512(TMP[0], TMP[1], TMP[2], TMP[3], TMP[16], TMP[17], TMP[18], TMP[19]); TMP[0] = _mm512_shuffle_epi8(TMP[0], M512(swapBytes)); TMP[1] = _mm512_shuffle_epi8(TMP[1], M512(swapBytes)); TMP[2] = _mm512_shuffle_epi8(TMP[2], M512(swapBytes)); TMP[3] = _mm512_shuffle_epi8(TMP[3], M512(swapBytes)); **_mm512_storeu_si512((__m512i*)loc_out[13], _mm512_xor_si512(TMP[1], TWEAK[13]));** _mm512_storeu_si512((__m512i*)loc_out[13], _mm512_xor_si512(TMP[1], TWEAK[13])); _mm512_storeu_si512((__m512i*)loc_out[14], _mm512_xor_si512(TMP[2], TWEAK[14])); _mm512_storeu_si512((__m512i*)loc_out[15], _mm512_xor_si512(TMP[3], TWEAK[15]));

It seems there is an issue at line 552.
_mm512_storeu_si512((__m512i*)loc_out[13], _mm512_xor_si512(TMP[1], TWEAK[13]));

it should be
__mm512_storeu_si512((__m512i*)loc_out[12], _mm512_xor_si512(TMP[0], TWEAK[12]));

[rcao8]

Thank you @Jingkai for posting. We will work on it.