Move Rijndael_UncheckedSetKey_POWER8 prior to GetUserKey call

Arg... GetUserKey was performing a 32-bit word reverse. It was part of the problem on little endian machines

1 files changed, 18 insertions, 3 deletions

diff --git a/rijndael-simd.cpp b/rijndael-simd.cpp
index 9f8102df..e558a26d 100644
--- a/rijndael-simd.cpp
+++ b/rijndael-simd.cpp
@@ -1070,7 +1070,13 @@ static const uint32_t s_rcon[3][4] = {
 /* Permute mask */

 CRYPTOPP_ALIGN_DATA(16)

 static const uint32_t s_mask[4] = {

+#if defined(IS_LITTLE_ENDIAN)

+	// 0x0d0e0f0c, 0x0d0e0f0c, 0x0d0e0f0c, 0x0d0e0f0c

+	// 0x01020300, 0x01020300, 0x01020300, 0x01020300

+	0x02010003, 0x02010003, 0x02010003, 0x02010003

+#else

 	0x0d0e0f0c, 0x0d0e0f0c, 0x0d0e0f0c, 0x0d0e0f0c

+#endif

 };

 

 static inline uint8x16_p8

@@ -1097,15 +1103,23 @@ Rijndael_Subkey_POWER8(uint8x16_p8 r1, const uint8x16_p8 r4, const uint8x16_p8 r
 	return r1;

 }

 

-void Rijndael_UncheckedSetKey_POWER8(word32* rk, size_t keyLen, const word32* rc,

+void Rijndael_UncheckedSetKey_POWER8(const byte* userKey, size_t keyLen, word32* rk, const word32* rc,

                                      const byte* Se, unsigned int rounds)

 {

 #if defined(IS_BIG_ENDIAN)

-	// Testing shows this is about 125 to 275 cycles faster.

+	// Testing shows this is about 150 to 350 cycles faster.

 	if (keyLen == 16)

 	{

+#if defined(IS_BIG_ENDIAN)

 		uint8_t* skptr = (uint8_t*)rk;

-		uint8x16_p8 r1 = (uint8x16_p8)VectorLoad((uint8_t*)skptr);

+		std::memcpy(rk, userKey, keyLen);

+#else

+		uint8_t* skptr = (uint8_t*)rk;

+		std::memcpy(rk, userKey, keyLen);

+		ReverseByteArrayLE(skptr);

+#endif

+

+		uint8x16_p8 r1 = (uint8x16_p8)VectorLoadKey(skptr);

 		uint8x16_p8 r4 = (uint8x16_p8)VectorLoadKey(s_rcon[0]);

 		uint8x16_p8 r5 = (uint8x16_p8)VectorLoadKey(s_mask);

 

@@ -1140,6 +1154,7 @@ void Rijndael_UncheckedSetKey_POWER8(word32* rk, size_t keyLen, const word32* rc
 	else

 #endif

 	{

+		GetUserKey(BIG_ENDIAN_ORDER, rk, keyLen/4, userKey, keyLen);

 		word32 *rk_saved = rk, temp;

 

 		// keySize: m_key allocates 4*(rounds+1) word32's.