arne
/
cryptopp
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Update comments

pull/548/head
Jeffrey Walton 2017-12-01 02:38:19 -05:00
parent a7fec9c0f6
commit b7ced67892
No known key found for this signature in database
GPG Key ID: B36AB348921B1838
1 changed files with 8 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
simon-simd.cpp
View File

@ -128,7 +128,7 @@ inline uint64x2_t SIMON128_f(const uint64x2_t& val)
inline void SIMON128_Enc_Block(uint8x16_t &block0, const word64 *subkeys, unsigned int rounds)
{
// Hack ahead... Rearrange the data for vectorization. It is easier to permute
// the data in SPECK128_Enc_Blocks then SPECK128_AdvancedProcessBlocks_NEON.
// the data in SIMON128_Enc_Blocks then SIMON128_AdvancedProcessBlocks_NEON.
// The zero block below is a "don't care". It is present so we can vectorize.
uint8x16_t block1 = {0};
uint64x2_t x1 = UnpackLow64<uint64x2_t>(block0, block1);
@ -170,7 +170,7 @@ inline void SIMON128_Enc_6_Blocks(uint8x16_t &block0, uint8x16_t &block1,
uint8x16_t &block5, const word64 *subkeys, unsigned int rounds)
{
// Hack ahead... Rearrange the data for vectorization. It is easier to permute
// the data in SPECK128_Enc_Blocks then SPECK128_AdvancedProcessBlocks_NEON.
// the data in SIMON128_Enc_Blocks then SIMON128_AdvancedProcessBlocks_NEON.
uint64x2_t x1 = UnpackLow64<uint64x2_t>(block0, block1);
uint64x2_t y1 = UnpackHigh64<uint64x2_t>(block0, block1);
uint64x2_t x2 = UnpackLow64<uint64x2_t>(block2, block3);
@ -238,7 +238,7 @@ inline void SIMON128_Enc_6_Blocks(uint8x16_t &block0, uint8x16_t &block1,
inline void SIMON128_Dec_Block(uint8x16_t &block0, const word64 *subkeys, unsigned int rounds)
{
// Hack ahead... Rearrange the data for vectorization. It is easier to permute
// the data in SPECK128_Dec_Blocks then SPECK128_AdvancedProcessBlocks_NEON.
// the data in SIMON128_Dec_Blocks then SIMON128_AdvancedProcessBlocks_NEON.
// The zero block below is a "don't care". It is present so we can vectorize.
uint8x16_t block1 = {0};
uint64x2_t x1 = UnpackLow64<uint64x2_t>(block0, block1);
@ -280,7 +280,7 @@ inline void SIMON128_Dec_6_Blocks(uint8x16_t &block0, uint8x16_t &block1,
uint8x16_t &block5, const word64 *subkeys, unsigned int rounds)
{
// Hack ahead... Rearrange the data for vectorization. It is easier to permute
// the data in SPECK128_Dec_Blocks then SPECK128_AdvancedProcessBlocks_NEON.
// the data in SIMON128_Dec_Blocks then SIMON128_AdvancedProcessBlocks_NEON.
uint64x2_t x1 = UnpackLow64<uint64x2_t>(block0, block1);
uint64x2_t y1 = UnpackHigh64<uint64x2_t>(block0, block1);
uint64x2_t x2 = UnpackLow64<uint64x2_t>(block2, block3);
@ -513,7 +513,7 @@ inline __m128i SIMON128_f(const __m128i& v)
inline void SIMON128_Enc_Block(__m128i &block0, const word64 *subkeys, unsigned int rounds)
{
// Hack ahead... Rearrange the data for vectorization. It is easier to permute
// the data in SPECK128_Enc_Blocks then SPECK128_AdvancedProcessBlocks_SSSE3.
// the data in SIMON128_Enc_Blocks then SIMON128_AdvancedProcessBlocks_SSSE3.
// The zero block below is a "don't care". It is present so we can vectorize.
__m128i block1 = _mm_setzero_si128();
__m128i x1 = _mm_unpacklo_epi64(block0, block1);
@ -558,7 +558,7 @@ inline void SIMON128_Enc_4_Blocks(__m128i &block0, __m128i &block1,
__m128i &block2, __m128i &block3, const word64 *subkeys, unsigned int rounds)
{
// Hack ahead... Rearrange the data for vectorization. It is easier to permute
// the data in SPECK128_Enc_Blocks then SPECK128_AdvancedProcessBlocks_SSSE3.
// the data in SIMON128_Enc_Blocks then SIMON128_AdvancedProcessBlocks_SSSE3.
__m128i x1 = _mm_unpacklo_epi64(block0, block1);
__m128i y1 = _mm_unpackhi_epi64(block0, block1);
__m128i x2 = _mm_unpacklo_epi64(block2, block3);
@ -615,7 +615,7 @@ inline void SIMON128_Enc_4_Blocks(__m128i &block0, __m128i &block1,
inline void SIMON128_Dec_Block(__m128i &block0, const word64 *subkeys, unsigned int rounds)
{
// Hack ahead... Rearrange the data for vectorization. It is easier to permute
// the data in SPECK128_Dec_Blocks then SPECK128_AdvancedProcessBlocks_SSSE3.
// the data in SIMON128_Dec_Blocks then SIMON128_AdvancedProcessBlocks_SSSE3.
// The zero block below is a "don't care". It is present so we can vectorize.
__m128i block1 = _mm_setzero_si128();
__m128i x1 = _mm_unpacklo_epi64(block0, block1);
@ -660,7 +660,7 @@ inline void SIMON128_Dec_4_Blocks(__m128i &block0, __m128i &block1,
__m128i &block2, __m128i &block3, const word64 *subkeys, unsigned int rounds)
{
// Hack ahead... Rearrange the data for vectorization. It is easier to permute
// the data in SPECK128_Dec_Blocks then SPECK128_AdvancedProcessBlocks_SSSE3.
// the data in SIMON128_Dec_Blocks then SIMON128_AdvancedProcessBlocks_SSSE3.
__m128i x1 = _mm_unpacklo_epi64(block0, block1);
__m128i y1 = _mm_unpackhi_epi64(block0, block1);
__m128i x2 = _mm_unpacklo_epi64(block2, block3);