Use Altivec aligned loads for Simon64

1 files changed, 9 insertions, 9 deletions

diff --git a/simon64_simd.cpp b/simon64_simd.cpp
index 21f9d129..d12041a5 100644
--- a/simon64_simd.cpp
+++ b/simon64_simd.cpp
@@ -520,7 +520,7 @@ using CryptoPP::uint32x4_p;
 using CryptoPP::VecAnd;

 using CryptoPP::VecXor;

 using CryptoPP::VecLoad;

-using CryptoPP::VecLoadBE;

+using CryptoPP::VecLoadAligned;

 using CryptoPP::VecPermute;

 

 // Rotate left by bit count

@@ -563,8 +563,8 @@ inline void SIMON64_Enc_Block(uint32x4_p &block0, uint32x4_p &block1,
     for (int i = 0; i < static_cast<int>(rounds & ~1)-1; i += 2)

     {

         // Round keys are pre-splated in forward direction

-        const uint32x4_p rk1 = VecLoad(subkeys+i*4);

-        const uint32x4_p rk2 = VecLoad(subkeys+i*4+4);

+        const uint32x4_p rk1 = VecLoadAligned(subkeys+i*4);

+        const uint32x4_p rk2 = VecLoadAligned(subkeys+i*4+4);

 

         y1 = VecXor(VecXor(y1, SIMON64_f(x1)), rk1);

         x1 = VecXor(VecXor(x1, SIMON64_f(y1)), rk2);

@@ -573,7 +573,7 @@ inline void SIMON64_Enc_Block(uint32x4_p &block0, uint32x4_p &block1,
     if (rounds & 1)

     {

         // Round keys are pre-splated in forward direction

-        const uint32x4_p rk = VecLoad(subkeys+rounds*4-4);

+        const uint32x4_p rk = VecLoadAligned(subkeys+rounds*4-4);

 

         y1 = VecXor(VecXor(y1, SIMON64_f(x1)), rk);

         std::swap(x1, y1);

@@ -673,8 +673,8 @@ inline void SIMON64_Enc_6_Blocks(uint32x4_p &block0, uint32x4_p &block1,
     for (int i = 0; i < static_cast<int>(rounds & ~1)-1; i += 2)

     {

         // Round keys are pre-splated in forward direction

-        const uint32x4_p rk1 = VecLoad(subkeys+i*4);

-        const uint32x4_p rk2 = VecLoad(subkeys+i*4+4);

+        const uint32x4_p rk1 = VecLoadAligned(subkeys+i*4);

+        const uint32x4_p rk2 = VecLoadAligned(subkeys+i*4+4);

 

         y1 = VecXor(VecXor(y1, SIMON64_f(x1)), rk1);

         y2 = VecXor(VecXor(y2, SIMON64_f(x2)), rk1);

@@ -688,7 +688,7 @@ inline void SIMON64_Enc_6_Blocks(uint32x4_p &block0, uint32x4_p &block1,
     if (rounds & 1)

     {

         // Round keys are pre-splated in forward direction

-        const uint32x4_p rk = VecLoad(subkeys+rounds*4-4);

+        const uint32x4_p rk = VecLoadAligned(subkeys+rounds*4-4);

 

         y1 = VecXor(VecXor(y1, SIMON64_f(x1)), rk);

         y2 = VecXor(VecXor(y2, SIMON64_f(x2)), rk);

@@ -756,8 +756,8 @@ inline void SIMON64_Dec_6_Blocks(uint32x4_p &block0, uint32x4_p &block1,
         const uint32x4_p rk2 = vec_splats(subkeys[i]);

 #else

         const uint8x16_p m = {0,1,2,3, 0,1,2,3, 0,1,2,3, 0,1,2,3};

-        uint32x4_p rk1 = VecLoad(subkeys+i+1);

-        uint32x4_p rk2 = VecLoad(subkeys+i);

+        uint32x4_p rk1 = VecLoadAligned(subkeys+i+1);

+        uint32x4_p rk2 = VecLoadAligned(subkeys+i);

         rk1 = VecPermute(rk1, rk1, m);

         rk2 = VecPermute(rk2, rk2, m);

 #endif