Use vec_xl and vec_xst in Rijndael set key

1 files changed, 20 insertions, 20 deletions

diff --git a/rijndael_simd.cpp b/rijndael_simd.cpp
index f8dbc15f..a08f17c0 100644
--- a/rijndael_simd.cpp
+++ b/rijndael_simd.cpp
@@ -153,7 +153,7 @@ bool CPU_ProbeAES()
 

 ANONYMOUS_NAMESPACE_BEGIN

 

-static inline void ARMV8_Enc_Block(uint64x2_t &data, const word32 *subkeys, unsigned int rounds)

+inline void ARMV8_Enc_Block(uint64x2_t &data, const word32 *subkeys, unsigned int rounds)

 {

     CRYPTOPP_ASSERT(subkeys);

     const byte *keys = reinterpret_cast<const byte*>(subkeys);

@@ -184,7 +184,7 @@ static inline void ARMV8_Enc_Block(uint64x2_t &data, const word32 *subkeys, unsi
     data = vreinterpretq_u64_u8(block);

 }

 

-static inline void ARMV8_Enc_6_Blocks(uint64x2_t &data0, uint64x2_t &data1,

+inline void ARMV8_Enc_6_Blocks(uint64x2_t &data0, uint64x2_t &data1,

     uint64x2_t &data2, uint64x2_t &data3, uint64x2_t &data4, uint64x2_t &data5,

     const word32 *subkeys, unsigned int rounds)

 {

@@ -247,7 +247,7 @@ static inline void ARMV8_Enc_6_Blocks(uint64x2_t &data0, uint64x2_t &data1,
     data5 = vreinterpretq_u64_u8(veorq_u8(block5, key));

 }

 

-static inline void ARMV8_Dec_Block(uint64x2_t &data, const word32 *subkeys, unsigned int rounds)

+inline void ARMV8_Dec_Block(uint64x2_t &data, const word32 *subkeys, unsigned int rounds)

 {

     CRYPTOPP_ASSERT(subkeys);

     const byte *keys = reinterpret_cast<const byte*>(subkeys);

@@ -278,7 +278,7 @@ static inline void ARMV8_Dec_Block(uint64x2_t &data, const word32 *subkeys, unsi
     data = vreinterpretq_u64_u8(block);

 }

 

-static inline void ARMV8_Dec_6_Blocks(uint64x2_t &data0, uint64x2_t &data1,

+inline void ARMV8_Dec_6_Blocks(uint64x2_t &data0, uint64x2_t &data1,

     uint64x2_t &data2, uint64x2_t &data3, uint64x2_t &data4, uint64x2_t &data5,

     const word32 *subkeys, unsigned int rounds)

 {

@@ -371,7 +371,7 @@ const word32 s_rconLE[] = {
     0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x1B, 0x36

 };

 

-static inline void AESNI_Enc_Block(__m128i &block, MAYBE_CONST word32 *subkeys, unsigned int rounds)

+inline void AESNI_Enc_Block(__m128i &block, MAYBE_CONST word32 *subkeys, unsigned int rounds)

 {

     const __m128i* skeys = reinterpret_cast<const __m128i*>(subkeys);

 

@@ -385,7 +385,7 @@ static inline void AESNI_Enc_Block(__m128i &block, MAYBE_CONST word32 *subkeys,
     block = _mm_aesenclast_si128(block, skeys[rounds]);

 }

 

-static inline void AESNI_Enc_4_Blocks(__m128i &block0, __m128i &block1, __m128i &block2, __m128i &block3,

+inline void AESNI_Enc_4_Blocks(__m128i &block0, __m128i &block1, __m128i &block2, __m128i &block3,

                                MAYBE_CONST word32 *subkeys, unsigned int rounds)

 {

     const __m128i* skeys = reinterpret_cast<const __m128i*>(subkeys);

@@ -410,7 +410,7 @@ static inline void AESNI_Enc_4_Blocks(__m128i &block0, __m128i &block1, __m128i
     block3 = _mm_aesenclast_si128(block3, rk);

 }

 

-static inline void AESNI_Dec_Block(__m128i &block, MAYBE_CONST word32 *subkeys, unsigned int rounds)

+inline void AESNI_Dec_Block(__m128i &block, MAYBE_CONST word32 *subkeys, unsigned int rounds)

 {

     const __m128i* skeys = reinterpret_cast<const __m128i*>(subkeys);

 

@@ -424,7 +424,7 @@ static inline void AESNI_Dec_Block(__m128i &block, MAYBE_CONST word32 *subkeys,
     block = _mm_aesdeclast_si128(block, skeys[rounds]);

 }

 

-static inline void AESNI_Dec_4_Blocks(__m128i &block0, __m128i &block1, __m128i &block2, __m128i &block3,

+inline void AESNI_Dec_4_Blocks(__m128i &block0, __m128i &block1, __m128i &block2, __m128i &block3,

                         MAYBE_CONST word32 *subkeys, unsigned int rounds)

 {

     const __m128i* skeys = reinterpret_cast<const __m128i*>(subkeys);

@@ -553,7 +553,7 @@ static const uint32_t s_rconBE[] = {
     0x1B000000, 0x36000000

 };

 

-static inline void POWER8_Enc_Block(uint32x4_p &block, const word32 *subkeys, unsigned int rounds)

+inline void POWER8_Enc_Block(uint32x4_p &block, const word32 *subkeys, unsigned int rounds)

 {

     CRYPTOPP_ASSERT(IsAlignedOn(subkeys, 16));

     const byte *keys = reinterpret_cast<const byte*>(subkeys);

@@ -571,7 +571,7 @@ static inline void POWER8_Enc_Block(uint32x4_p &block, const word32 *subkeys, un
     block = VecEncryptLast(block, VecLoadAligned(rounds*16, keys));

 }

 

-static inline void POWER8_Enc_6_Blocks(uint32x4_p &block0, uint32x4_p &block1,

+inline void POWER8_Enc_6_Blocks(uint32x4_p &block0, uint32x4_p &block1,

             uint32x4_p &block2, uint32x4_p &block3, uint32x4_p &block4,

             uint32x4_p &block5, const word32 *subkeys, unsigned int rounds)

 {

@@ -606,7 +606,7 @@ static inline void POWER8_Enc_6_Blocks(uint32x4_p &block0, uint32x4_p &block1,
     block5 = VecEncryptLast(block5, k);

 }

 

-static inline void POWER8_Dec_Block(uint32x4_p &block, const word32 *subkeys, unsigned int rounds)

+inline void POWER8_Dec_Block(uint32x4_p &block, const word32 *subkeys, unsigned int rounds)

 {

     CRYPTOPP_ASSERT(IsAlignedOn(subkeys, 16));

     const byte *keys = reinterpret_cast<const byte*>(subkeys);

@@ -624,7 +624,7 @@ static inline void POWER8_Dec_Block(uint32x4_p &block, const word32 *subkeys, un
     block = VecDecryptLast(block, VecLoadAligned(0, keys));

 }

 

-static inline void POWER8_Dec_6_Blocks(uint32x4_p &block0, uint32x4_p &block1,

+inline void POWER8_Dec_6_Blocks(uint32x4_p &block0, uint32x4_p &block1,

             uint32x4_p &block2, uint32x4_p &block3, uint32x4_p &block4,

             uint32x4_p &block5, const word32 *subkeys, unsigned int rounds)

 {

@@ -704,22 +704,22 @@ void Rijndael_UncheckedSetKey_POWER8(const byte* userKey, size_t keyLen, word32*
 

 #if (CRYPTOPP_LITTLE_ENDIAN)

     rkey = rk;

-    const uint8x16_p mask = ((uint8x16_p){12,13,14,15, 8,9,10,11, 4,5,6,7, 0,1,2,3});

-    const uint8x16_p zero = {0};

+    const uint8x16_p mask = {12,13,14,15, 8,9,10,11, 4,5,6,7, 0,1,2,3};

+    const uint32x4_p zero = {0,0,0,0};

 

     unsigned int i=0;

     for (i=0; i<rounds; i+=2, rkey+=8)

     {

-        const uint8x16_p d1 = vec_vsx_ld( 0, (uint8_t*)rkey);

-        const uint8x16_p d2 = vec_vsx_ld(16, (uint8_t*)rkey);

-        vec_vsx_st(VecPermute(d1, zero, mask),  0, (uint8_t*)rkey);

-        vec_vsx_st(VecPermute(d2, zero, mask), 16, (uint8_t*)rkey);

+        const uint32x4_p d1 = vec_xl( 0, rkey);

+        const uint32x4_p d2 = vec_xl(16, rkey);

+        vec_xst(VecPermute(d1, zero, mask),  0, rkey);

+        vec_xst(VecPermute(d2, zero, mask), 16, rkey);

     }

 

     for ( ; i<rounds+1; i++, rkey+=4)

     {

-        const uint8x16_p d = vec_vsx_ld( 0, (uint8_t*)rkey);

-        vec_vsx_st(VecPermute(d, zero, mask),  0, (uint8_t*)rkey);

+        const uint32x4_p d = vec_xl( 0, rkey);

+        vec_xst(VecPermute(d, zero, mask),  0, rkey);

     }

 #endif

 }