knot

knot/Implementations/crypto_aead/knot128v2/armcortexm_2/auxFormat.c

 #include"auxFormat.h"
 #include"auxFormat.h"
 
 
-void packU96FormatToThreePacket(u32 *out, u8 *in) {
-	u32 t0 = U32BIG(((u32*)in)[0]), t1 = U32BIG(((u32*)in)[1]), t2 = U32BIG(((u32*)in)[2]);	\
-	puckU32ToThree_3(t0);	\
-	puckU32ToThree_3(t1);	\
-	puckU32ToThree_3(t2);	\
-
-out[0] = ((((t2 >> 21) & 0x400) | (((t2 >> 18) & 0x300)) | ((t2 & 0xff00) >> 8)) << 21) |
-	     ((((t1 >> 20) & 0x400) | ((t1 >> 16) & 0x300) | (t1 & 0xff))        << 10) |
-		 (((t0 >> 20) & 0x300) | ((t0 & 0xff0000) >> 16));
-out[1] = ((((t2 >> 20) & 0x400) | ((t2 >> 16) & 0x300)   | (t2 & 0xff))          << 21) |
-	     ((((t1 >> 20) & 0x300) | ((t1 & 0xff0000) >> 16))                   << 11) |
-	     (((t0 >> 21) & 0x400) | ((t0 >> 18) & 0x300) | ((t0 & 0xff00) >> 8));
-out[2] = ((((t2 >> 20) & 0x300) | ((t2 & 0xff0000) >> 16))                       << 22) |
-	     ((((t1 >> 21) & 0x400) | ((t1 >> 18) & 0x300)|((t1 & 0xff00) >> 8)) << 11) |
-	     (((t0 >> 20) & 0x400) | ((t0 >> 16) & 0x300) | (t0 & 0xff));
+void packU96FormatToThreePacket(u32 * out, u8 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t1=U32BIG(((u32*)in)[0]);
+	temp0[0] = t1; temp0[1] = t1 >> 1; temp0[2] = t1>> 2;
+	puckU32ToThree_1(temp0[0]);
+	puckU32ToThree_1(temp0[1]);
+	puckU32ToThree_1(temp0[2]);
+	t1=U32BIG(((u32*)in)[1]);
+	temp1[0] = t1; temp1[1] = t1>>1; temp1[2] = t1 >> 2;
+	puckU32ToThree_1(temp1[0]);
+	puckU32ToThree_1(temp1[1]);
+	puckU32ToThree_1(temp1[2]);
+	t1=U32BIG(((u32*)in)[2]);
+	temp2[0] = t1; temp2[1] =t1 >> 1; temp2[2] = t1>> 2;
+	puckU32ToThree_1(temp2[0]);
+	puckU32ToThree_1(temp2[1]);
+	puckU32ToThree_1(temp2[2]);
+	out[0] = (temp2[1]<<21)	|(temp1[0]<<10)	|temp0[2];
+	out[1] = (temp2[0] << 21) | (temp1[2] << 11) | temp0[1];
+	out[2] = (temp2[2] << 22) | (temp1[1] << 11) | temp0[0];
 }
 }
-void unpackU96FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 } ;\
-u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) |		 \
-((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) |		 \
-((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10) | ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7) |		 \
-((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5) | ((t2 & 0x7f800) >> 3) |		 \
-((t0 & 0x3fc00) >> 10);		 \
-t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1) | ((t2 & 0xc0000000) >> 2) |		 \
-((t0 & 0x60000000) >> 3) | ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6) |		 \
-((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);		 \
-unpuckU32ToThree_3(t[0]);		 \
-unpuckU32ToThree_3(t[1]);		 \
-unpuckU32ToThree_3(t[2]);		 \
-memcpy(out, t, 12 * sizeof(unsigned char));		 \
+void unpackU96FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t[3] = { 0 };
+	u32 t0=in[0] ;
+	u32 t1=in[1] ;
+	u32 t2=in[2] ;
+	temp0[0] = t2 & 0x7ff;
+	temp0[1] = t1 & 0x7ff;
+	temp0[2] = t0 & 0x3ff;
+	temp1[0] = (t0>>10) & 0x7ff;
+	temp1[1] = (t2 >>11 ) & 0x7ff;
+	temp1[2] = (t1 >> 11) & 0x3ff;
+	temp2[0] = t1 >> 21;
+	temp2[1] = t0 >> 21;
+	temp2[2] = t2 >> 22;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;
+	unpuckU32ToThree_1(temp1[0]);
+	unpuckU32ToThree_1(temp1[1]);
+	unpuckU32ToThree_1(temp1[2]);
+	t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;
+	unpuckU32ToThree_1(temp2[0]);
+	unpuckU32ToThree_1(temp2[1]);
+	unpuckU32ToThree_1(temp2[2]);
+	t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;
+	memcpy(out, t, 12 * sizeof(unsigned char));
 }
 }
+
 void P384(unsigned int *s, unsigned char *round, unsigned char lunNum) {
 void P384(unsigned int *s, unsigned char *round, unsigned char lunNum) {
 	u32 rci,t1,t2;
 	u32 rci,t1,t2;
 	unsigned char  i;
 	unsigned char  i;

knot/Implementations/crypto_aead/knot128v2/armcortexm_2/auxFormat.h

@@ -31,24 +31,21 @@ void packU96FormatToThreePacket(u32 * out, u8 * in);
 void unpackU96FormatToThreePacket(u8 * out, u32 * in);
 void unpackU96FormatToThreePacket(u8 * out, u32 * in);
 void P384(unsigned int *s, unsigned char *round, unsigned char lunNum);
 void P384(unsigned int *s, unsigned char *round, unsigned char lunNum);
 
 
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }
+
 unsigned char  constant7Format[80];
 unsigned char  constant7Format[80];
 
 
 #define P384_ARC_SC1(rci,S2,S3,S4) \
 #define P384_ARC_SC1(rci,S2,S3,S4) \

knot/Implementations/crypto_aead/knot128v2/armcortexm_3/auxFormat.h

@@ -52,6 +52,20 @@ do { \
 			s[4] ^= dataFormat[4];          \
 			s[4] ^= dataFormat[4];          \
 			s[5] ^= dataFormat[5];          \
 			s[5] ^= dataFormat[5];          \
 } while (0)
 } while (0)
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
+}
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
+}
 
 
 
 
 #define puckU32ToThree_3(lo){\
 #define puckU32ToThree_3(lo){\
@@ -74,36 +88,51 @@ r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
 }
 }
 
 
 #define packU96FormatToThreePacket( out,  in) {               \
 #define packU96FormatToThreePacket( out,  in) {               \
-		u32  t0 = U32BIG(((u32*)in)[0]), t1 = U32BIG(((u32*)in)[1]), t2 = U32BIG(((u32*)in)[2]);	\
-		puckU32ToThree_3(t0);	\
-		puckU32ToThree_3(t1);	\
-		puckU32ToThree_3(t2);	\
-		out[0] = ((((t2 >> 21) & 0x400) | (((t2 >> 18) & 0x300)) | ((t2 & 0xff00) >> 8)) << 21) |										\
-			     ((((t1 >> 20) & 0x400) | ((t1 >> 16) & 0x300) | (t1 & 0xff))        << 10) |									\
-				 (((t0 >> 20) & 0x300) | ((t0 & 0xff0000) >> 16));								\
-		out[1] = ((((t2 >> 20) & 0x400) | ((t2 >> 16) & 0x300)   | (t2 & 0xff))          << 21) |										\
-			     ((((t1 >> 20) & 0x300) | ((t1 & 0xff0000) >> 16))                   << 11) |									\
-			     (((t0 >> 21) & 0x400) | ((t0 >> 18) & 0x300) | ((t0 & 0xff00) >> 8));									\
-		out[2] = ((((t2 >> 20) & 0x300) | ((t2 & 0xff0000) >> 16))                       << 22) |										\
-			     ((((t1 >> 21) & 0x400) | ((t1 >> 18) & 0x300)|((t1 & 0xff00) >> 8)) << 11) |									\
-			     (((t0 >> 20) & 0x400) | ((t0 >> 16) & 0x300) | (t0 & 0xff));									\
+		t1=U32BIG(((u32*)in)[0]);			\
+		temp0[0] = t1; temp0[1] = t1 >> 1; temp0[2] = t1>> 2;			\
+		puckU32ToThree_1(temp0[0]);			\
+		puckU32ToThree_1(temp0[1]);			\
+		puckU32ToThree_1(temp0[2]);			\
+		t1=U32BIG(((u32*)in)[1]);			\
+		temp1[0] = t1; temp1[1] = t1>>1; temp1[2] = t1 >> 2;			\
+		puckU32ToThree_1(temp1[0]);			\
+		puckU32ToThree_1(temp1[1]);			\
+		puckU32ToThree_1(temp1[2]);			\
+		t1=U32BIG(((u32*)in)[2]);			\
+		temp2[0] = t1; temp2[1] =t1 >> 1; temp2[2] = t1>> 2;			\
+		puckU32ToThree_1(temp2[0]);			\
+		puckU32ToThree_1(temp2[1]);			\
+		puckU32ToThree_1(temp2[2]);			\
+		out[0] = (temp2[1]<<21)	|(temp1[0]<<10)	|temp0[2];	\
+		out[1] = (temp2[0] << 21) | (temp1[2] << 11) | temp0[1];			\
+		out[2] = (temp2[2] << 22) | (temp1[1] << 11) | temp0[0];			\
 }
 }
 
 
 #define unpackU96FormatToThreePacket( out, in) {\
 #define unpackU96FormatToThreePacket( out, in) {\
-		u32 t[3] = { 0 };\
-	u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-	t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) |		 \
-	((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) |		 \
-	((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-	t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10) | ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7) |		 \
-	((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5) | ((t2 & 0x7f800) >> 3) |		 \
-	((t0 & 0x3fc00) >> 10);		 \
-	t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1) | ((t2 & 0xc0000000) >> 2) |		 \
-	((t0 & 0x60000000) >> 3) | ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6) |		 \
-	((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);		 \
-	unpuckU32ToThree_3(t[0]);		 \
-	unpuckU32ToThree_3(t[1]);		 \
-	unpuckU32ToThree_3(t[2]);		 \
+		t3=in[0] ;			\
+		t1=in[1] ;			\
+		t2=in[2] ;			\
+		temp0[0] = t2 & 0x7ff;			\
+		temp0[1] = t1 & 0x7ff;			\
+		temp0[2] = t3 & 0x3ff;			\
+		temp1[0] = (t3>>10) & 0x7ff;			\
+		temp1[1] = (t2 >>11 ) & 0x7ff;			\
+		temp1[2] = (t1 >> 11) & 0x3ff;			\
+		temp2[0] = t1 >> 21;			\
+		temp2[1] = t3 >> 21;			\
+		temp2[2] = t2 >> 22;			\
+		unpuckU32ToThree_1(temp0[0]);			\
+		unpuckU32ToThree_1(temp0[1]);			\
+		unpuckU32ToThree_1(temp0[2]);			\
+		t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;			\
+		unpuckU32ToThree_1(temp1[0]);			\
+		unpuckU32ToThree_1(temp1[1]);			\
+		unpuckU32ToThree_1(temp1[2]);			\
+		t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;			\
+		unpuckU32ToThree_1(temp2[0]);			\
+		unpuckU32ToThree_1(temp2[1]);			\
+		unpuckU32ToThree_1(temp2[2]);			\
+		t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;			\
 		memcpy(out, t, 12 * sizeof(unsigned char));			\
 		memcpy(out, t, 12 * sizeof(unsigned char));			\
 }
 }
 
 

knot/Implementations/crypto_aead/knot128v2/armcortexm_3/encrypt.c

@@ -34,6 +34,10 @@ int crypto_aead_encrypt(unsigned char *c, unsigned long long *clen,
 	u32 dataFormat[6] = { 0 };
 	u32 dataFormat[6] = { 0 };
 	u32 s_temp[12] = { 0 };
 	u32 s_temp[12] = { 0 };
 	u32 t1, t2, t3, t5, t6, t8, t9, t11;
 	u32 t1, t2, t3, t5, t6, t8, t9, t11;
+	u32 t[3] = { 0 };
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
 	u8 tempU8[24] = { 0 };
 	u8 tempU8[24] = { 0 };
 	*clen = mlen + CRYPTO_ABYTES;
 	*clen = mlen + CRYPTO_ABYTES;
 	// initialization
 	// initialization
@@ -112,6 +116,10 @@ int crypto_aead_decrypt(unsigned char *m, unsigned long long *mlen,
 	u8 tempData[24] = { 0 };
 	u8 tempData[24] = { 0 };
 	u8 tempU8[24] = { 0 };
 	u8 tempU8[24] = { 0 };
 	u32 t1, t2, t3, t5, t6, t8, t9, t11;
 	u32 t1, t2, t3, t5, t6, t8, t9, t11;
+	u32 t[3] = { 0 };
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
 	*mlen = clen - CRYPTO_ABYTES;
 	*mlen = clen - CRYPTO_ABYTES;
 	if (clen < CRYPTO_ABYTES)
 	if (clen < CRYPTO_ABYTES)
 		return -1;
 		return -1;

knot/Implementations/crypto_aead/knot128v2/armcortexm_4/auxFormat.c

 #include"auxFormat.h"
 #include"auxFormat.h"
-void packU96FormatToThreePacket(u32 *out, u8 *in) {
-	u32  t0 = U32BIG(((u32*)in)[0]), t1 = U32BIG(((u32*)in)[1]), t2 = U32BIG(((u32*)in)[2]);	\
-	puckU32ToThree_3(t0);	\
-	puckU32ToThree_3(t1);	\
-	puckU32ToThree_3(t2);	\
-	out[0] = ((((t2 >> 21) & 0x400) | (((t2 >> 18) & 0x300)) | ((t2 & 0xff00) >> 8)) << 21) |										\
-		     ((((t1 >> 20) & 0x400) | ((t1 >> 16) & 0x300) | (t1 & 0xff))        << 10) |									\
-			 (((t0 >> 20) & 0x300) | ((t0 & 0xff0000) >> 16));								\
-	out[1] = ((((t2 >> 20) & 0x400) | ((t2 >> 16) & 0x300)   | (t2 & 0xff))          << 21) |										\
-		     ((((t1 >> 20) & 0x300) | ((t1 & 0xff0000) >> 16))                   << 11) |									\
-		     (((t0 >> 21) & 0x400) | ((t0 >> 18) & 0x300) | ((t0 & 0xff00) >> 8));									\
-	out[2] = ((((t2 >> 20) & 0x300) | ((t2 & 0xff0000) >> 16))                       << 22) |										\
-		     ((((t1 >> 21) & 0x400) | ((t1 >> 18) & 0x300)|((t1 & 0xff00) >> 8)) << 11) |									\
-		     (((t0 >> 20) & 0x400) | ((t0 >> 16) & 0x300) | (t0 & 0xff));									\
-
+void packU96FormatToThreePacket(u32 * out, u8 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t1=U32BIG(((u32*)in)[0]);
+	temp0[0] = t1; temp0[1] = t1 >> 1; temp0[2] = t1>> 2;
+	puckU32ToThree_1(temp0[0]);
+	puckU32ToThree_1(temp0[1]);
+	puckU32ToThree_1(temp0[2]);
+	t1=U32BIG(((u32*)in)[1]);
+	temp1[0] = t1; temp1[1] = t1>>1; temp1[2] = t1 >> 2;
+	puckU32ToThree_1(temp1[0]);
+	puckU32ToThree_1(temp1[1]);
+	puckU32ToThree_1(temp1[2]);
+	t1=U32BIG(((u32*)in)[2]);
+	temp2[0] = t1; temp2[1] =t1 >> 1; temp2[2] = t1>> 2;
+	puckU32ToThree_1(temp2[0]);
+	puckU32ToThree_1(temp2[1]);
+	puckU32ToThree_1(temp2[2]);
+	out[0] = (temp2[1]<<21)	|(temp1[0]<<10)	|temp0[2];
+	out[1] = (temp2[0] << 21) | (temp1[2] << 11) | temp0[1];
+	out[2] = (temp2[2] << 22) | (temp1[1] << 11) | temp0[0];
 }
 }
-void unpackU96FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 } ;\
-u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) |		 \
-((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) |		 \
-((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10) | ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7) |		 \
-((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5) | ((t2 & 0x7f800) >> 3) |		 \
-((t0 & 0x3fc00) >> 10);		 \
-t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1) | ((t2 & 0xc0000000) >> 2) |		 \
-((t0 & 0x60000000) >> 3) | ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6) |		 \
-((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);		 \
-unpuckU32ToThree_3(t[0]);		 \
-unpuckU32ToThree_3(t[1]);		 \
-unpuckU32ToThree_3(t[2]);		 \
-memcpy(out, t, 12 * sizeof(unsigned char));		 \
+void unpackU96FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t[3] = { 0 };
+	u32 t0=in[0] ;
+	u32 t1=in[1] ;
+	u32 t2=in[2] ;
+	temp0[0] = t2 & 0x7ff;
+	temp0[1] = t1 & 0x7ff;
+	temp0[2] = t0 & 0x3ff;
+	temp1[0] = (t0>>10) & 0x7ff;
+	temp1[1] = (t2 >>11 ) & 0x7ff;
+	temp1[2] = (t1 >> 11) & 0x3ff;
+	temp2[0] = t1 >> 21;
+	temp2[1] = t0 >> 21;
+	temp2[2] = t2 >> 22;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;
+	unpuckU32ToThree_1(temp1[0]);
+	unpuckU32ToThree_1(temp1[1]);
+	unpuckU32ToThree_1(temp1[2]);
+	t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;
+	unpuckU32ToThree_1(temp2[0]);
+	unpuckU32ToThree_1(temp2[1]);
+	unpuckU32ToThree_1(temp2[2]);
+	t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;
+	memcpy(out, t, 12 * sizeof(unsigned char));
 }
 }
 void P384(unsigned int *s, unsigned char *round, unsigned char lunNum) {
 void P384(unsigned int *s, unsigned char *round, unsigned char lunNum) {
 	u32 s_temp[12] = { 0 };
 	u32 s_temp[12] = { 0 };

knot/Implementations/crypto_aead/knot128v2/armcortexm_4/auxFormat.h

@@ -46,22 +46,18 @@ void packU96FormatToThreePacket(u32 * out, u8 * in);
 void unpackU96FormatToThreePacket(u8 * out, u32 * in);
 void unpackU96FormatToThreePacket(u8 * out, u32 * in);
 
 
 unsigned char  constant7Format[80];
 unsigned char  constant7Format[80];
-
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }
+

knot/Implementations/crypto_aead/knot128v2/armcortexm_5/auxFormat.c

@@ -17,37 +17,58 @@ void ROUND384_Three(unsigned int *s, unsigned char *c, int lunnum) {
 		t++;
 		t++;
 	}
 	}
 }
 }
-void packU96FormatToThreePacket(u32 *out, u8 *in) {
-	u32 t0 = U32BIG(((u32*)in)[0]), t1 = U32BIG(((u32*)in)[1]), t2 = U32BIG(((u32*)in)[2]);	\
-	puckU32ToThree_3(t0);	\
-	puckU32ToThree_3(t1);	\
-	puckU32ToThree_3(t2);	\
-	out[0] = ((((t2 >> 21) & 0x400) | (((t2 >> 18) & 0x300)) | ((t2 & 0xff00) >> 8)) << 21) |										\
-		     ((((t1 >> 20) & 0x400) | ((t1 >> 16) & 0x300) | (t1 & 0xff))        << 10) |									\
-			 (((t0 >> 20) & 0x300) | ((t0 & 0xff0000) >> 16));								\
-	out[1] = ((((t2 >> 20) & 0x400) | ((t2 >> 16) & 0x300)   | (t2 & 0xff))          << 21) |										\
-		     ((((t1 >> 20) & 0x300) | ((t1 & 0xff0000) >> 16))                   << 11) |									\
-		     (((t0 >> 21) & 0x400) | ((t0 >> 18) & 0x300) | ((t0 & 0xff00) >> 8));									\
-	out[2] = ((((t2 >> 20) & 0x300) | ((t2 & 0xff0000) >> 16))                       << 22) |										\
-		     ((((t1 >> 21) & 0x400) | ((t1 >> 18) & 0x300)|((t1 & 0xff00) >> 8)) << 11) |									\
-		     (((t0 >> 20) & 0x400) | ((t0 >> 16) & 0x300) | (t0 & 0xff));									\
-
+void packU96FormatToThreePacket(u32 * out, u8 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t1=U32BIG(((u32*)in)[0]);
+	temp0[0] = t1; temp0[1] = t1 >> 1; temp0[2] = t1>> 2;
+	puckU32ToThree_1(temp0[0]);
+	puckU32ToThree_1(temp0[1]);
+	puckU32ToThree_1(temp0[2]);
+	t1=U32BIG(((u32*)in)[1]);
+	temp1[0] = t1; temp1[1] = t1>>1; temp1[2] = t1 >> 2;
+	puckU32ToThree_1(temp1[0]);
+	puckU32ToThree_1(temp1[1]);
+	puckU32ToThree_1(temp1[2]);
+	t1=U32BIG(((u32*)in)[2]);
+	temp2[0] = t1; temp2[1] =t1 >> 1; temp2[2] = t1>> 2;
+	puckU32ToThree_1(temp2[0]);
+	puckU32ToThree_1(temp2[1]);
+	puckU32ToThree_1(temp2[2]);
+	out[0] = (temp2[1]<<21)	|(temp1[0]<<10)	|temp0[2];
+	out[1] = (temp2[0] << 21) | (temp1[2] << 11) | temp0[1];
+	out[2] = (temp2[2] << 22) | (temp1[1] << 11) | temp0[0];
 }
 }
-void unpackU96FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 } ;\
-u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) |		 \
-((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) |		 \
-((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10) | ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7) |		 \
-((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5) | ((t2 & 0x7f800) >> 3) |		 \
-((t0 & 0x3fc00) >> 10);		 \
-t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1) | ((t2 & 0xc0000000) >> 2) |		 \
-((t0 & 0x60000000) >> 3) | ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6) |		 \
-((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);		 \
-unpuckU32ToThree_3(t[0]);		 \
-unpuckU32ToThree_3(t[1]);		 \
-unpuckU32ToThree_3(t[2]);		 \
-memcpy(out, t, 12 * sizeof(unsigned char));		 \
+void unpackU96FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t[3] = { 0 };
+	u32 t0=in[0] ;
+	u32 t1=in[1] ;
+	u32 t2=in[2] ;
+	temp0[0] = t2 & 0x7ff;
+	temp0[1] = t1 & 0x7ff;
+	temp0[2] = t0 & 0x3ff;
+	temp1[0] = (t0>>10) & 0x7ff;
+	temp1[1] = (t2 >>11 ) & 0x7ff;
+	temp1[2] = (t1 >> 11) & 0x3ff;
+	temp2[0] = t1 >> 21;
+	temp2[1] = t0 >> 21;
+	temp2[2] = t2 >> 22;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;
+	unpuckU32ToThree_1(temp1[0]);
+	unpuckU32ToThree_1(temp1[1]);
+	unpuckU32ToThree_1(temp1[2]);
+	t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;
+	unpuckU32ToThree_1(temp2[0]);
+	unpuckU32ToThree_1(temp2[1]);
+	unpuckU32ToThree_1(temp2[2]);
+	t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;
+	memcpy(out, t, 12 * sizeof(unsigned char));
 }
 }
 
 

knot/Implementations/crypto_aead/knot128v2/armcortexm_5/auxFormat.h

@@ -13,25 +13,22 @@ void ROUND384_Three(unsigned int *s, unsigned char  *c,int lunnum);
 
 
 
 
 
 
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-		u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }
 
 
+
 #define ARR_SIZE(a) (sizeof((a))/sizeof((a[0])))
 #define ARR_SIZE(a) (sizeof((a))/sizeof((a[0])))
 //Processing_Data:
 //Processing_Data:
 #define Processing_Data(data) \
 #define Processing_Data(data) \

knot/Implementations/crypto_aead/knot128v2/armcortexm_6/auxFormat.c

 #include"auxFormat.h"
 #include"auxFormat.h"
 
 
-
-void packU96FormatToThreePacket(u32 *out, u8 *in) {
-	u32  t0 = U32BIG(((u32*)in)[0]), t1 = U32BIG(((u32*)in)[1]), t2 = U32BIG(((u32*)in)[2]);	\
-	puckU32ToThree_3(t0);	\
-	puckU32ToThree_3(t1);	\
-	puckU32ToThree_3(t2);	\
-	out[0] = ((((t2 >> 21) & 0x400) | (((t2 >> 18) & 0x300)) | ((t2 & 0xff00) >> 8)) << 21) |										\
-		     ((((t1 >> 20) & 0x400) | ((t1 >> 16) & 0x300) | (t1 & 0xff))        << 10) |									\
-			 (((t0 >> 20) & 0x300) | ((t0 & 0xff0000) >> 16));								\
-	out[1] = ((((t2 >> 20) & 0x400) | ((t2 >> 16) & 0x300)   | (t2 & 0xff))          << 21) |										\
-		     ((((t1 >> 20) & 0x300) | ((t1 & 0xff0000) >> 16))                   << 11) |									\
-		     (((t0 >> 21) & 0x400) | ((t0 >> 18) & 0x300) | ((t0 & 0xff00) >> 8));									\
-	out[2] = ((((t2 >> 20) & 0x300) | ((t2 & 0xff0000) >> 16))                       << 22) |										\
-		     ((((t1 >> 21) & 0x400) | ((t1 >> 18) & 0x300)|((t1 & 0xff00) >> 8)) << 11) |									\
-		     (((t0 >> 20) & 0x400) | ((t0 >> 16) & 0x300) | (t0 & 0xff));									\
-
+void packU96FormatToThreePacket(u32 * out, u8 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t1=U32BIG(((u32*)in)[0]);
+	temp0[0] = t1; temp0[1] = t1 >> 1; temp0[2] = t1>> 2;
+	puckU32ToThree_1(temp0[0]);
+	puckU32ToThree_1(temp0[1]);
+	puckU32ToThree_1(temp0[2]);
+	t1=U32BIG(((u32*)in)[1]);
+	temp1[0] = t1; temp1[1] = t1>>1; temp1[2] = t1 >> 2;
+	puckU32ToThree_1(temp1[0]);
+	puckU32ToThree_1(temp1[1]);
+	puckU32ToThree_1(temp1[2]);
+	t1=U32BIG(((u32*)in)[2]);
+	temp2[0] = t1; temp2[1] =t1 >> 1; temp2[2] = t1>> 2;
+	puckU32ToThree_1(temp2[0]);
+	puckU32ToThree_1(temp2[1]);
+	puckU32ToThree_1(temp2[2]);
+	out[0] = (temp2[1]<<21)	|(temp1[0]<<10)	|temp0[2];
+	out[1] = (temp2[0] << 21) | (temp1[2] << 11) | temp0[1];
+	out[2] = (temp2[2] << 22) | (temp1[1] << 11) | temp0[0];
 }
 }
-void unpackU96FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 } ;\
-u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) |		 \
-((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) |		 \
-((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10) | ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7) |		 \
-((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5) | ((t2 & 0x7f800) >> 3) |		 \
-((t0 & 0x3fc00) >> 10);		 \
-t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1) | ((t2 & 0xc0000000) >> 2) |		 \
-((t0 & 0x60000000) >> 3) | ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6) |		 \
-((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);		 \
-unpuckU32ToThree_3(t[0]);		 \
-unpuckU32ToThree_3(t[1]);		 \
-unpuckU32ToThree_3(t[2]);		 \
-memcpy(out, t, 12 * sizeof(unsigned char));		 \
+void unpackU96FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t[3] = { 0 };
+	u32 t0=in[0] ;
+	u32 t1=in[1] ;
+	u32 t2=in[2] ;
+	temp0[0] = t2 & 0x7ff;
+	temp0[1] = t1 & 0x7ff;
+	temp0[2] = t0 & 0x3ff;
+	temp1[0] = (t0>>10) & 0x7ff;
+	temp1[1] = (t2 >>11 ) & 0x7ff;
+	temp1[2] = (t1 >> 11) & 0x3ff;
+	temp2[0] = t1 >> 21;
+	temp2[1] = t0 >> 21;
+	temp2[2] = t2 >> 22;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;
+	unpuckU32ToThree_1(temp1[0]);
+	unpuckU32ToThree_1(temp1[1]);
+	unpuckU32ToThree_1(temp1[2]);
+	t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;
+	unpuckU32ToThree_1(temp2[0]);
+	unpuckU32ToThree_1(temp2[1]);
+	unpuckU32ToThree_1(temp2[2]);
+	t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;
+	memcpy(out, t, 12 * sizeof(unsigned char));
 }
 }
-
 void ROUND384_Three(unsigned int *s, unsigned char *c, int lunnum) {
 void ROUND384_Three(unsigned int *s, unsigned char *c, int lunnum) {
 	unsigned int t, t1, t2;
 	unsigned int t, t1, t2;
 	u32 rci, temp;
 	u32 rci, temp;

knot/Implementations/crypto_aead/knot128v2/armcortexm_6/auxFormat.h

@@ -11,25 +11,22 @@ typedef unsigned long long u64;
 
 
 void ROUND384_Three(unsigned int *s, unsigned char  *c,int lunnum);
 void ROUND384_Three(unsigned int *s, unsigned char  *c,int lunnum);
 
 
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }
 
 
+
 #define ARR_SIZE(a) (sizeof((a))/sizeof((a[0])))
 #define ARR_SIZE(a) (sizeof((a))/sizeof((a[0])))
 //Processing_Data:
 //Processing_Data:
 #define Processing_Data(data) \
 #define Processing_Data(data) \

knot/Implementations/crypto_aead/knot192/armcortexm_2/auxFormat.c

@@ -10,42 +10,61 @@ void P384(unsigned int *s, unsigned char *round, unsigned char lunNum) {
 		P384_SR();\
 		P384_SR();\
 	}
 	}
 }
 }
-
-
-
-void packU96FormatToThreePacket(u32 *out, u8 *in) {
-	u32 t0 = U32BIG(((u32*)in)[0]), t1 = U32BIG(((u32*)in)[1]), t2 = U32BIG(((u32*)in)[2]);	\
-	puckU32ToThree_3(t0);	\
-	puckU32ToThree_3(t1);	\
-	puckU32ToThree_3(t2);	\
-
-	out[0] = ((((t2 >> 21) & 0x400) | (((t2 >> 18) & 0x300)) | ((t2 & 0xff00) >> 8)) << 21) |
-		     ((((t1 >> 20) & 0x400) | ((t1 >> 16) & 0x300) | (t1 & 0xff))        << 10) |
-			 (((t0 >> 20) & 0x300) | ((t0 & 0xff0000) >> 16));
-	out[1] = ((((t2 >> 20) & 0x400) | ((t2 >> 16) & 0x300)   | (t2 & 0xff))          << 21) |
-		     ((((t1 >> 20) & 0x300) | ((t1 & 0xff0000) >> 16))                   << 11) |
-		     (((t0 >> 21) & 0x400) | ((t0 >> 18) & 0x300) | ((t0 & 0xff00) >> 8));
-	out[2] = ((((t2 >> 20) & 0x300) | ((t2 & 0xff0000) >> 16))                       << 22) |
-		     ((((t1 >> 21) & 0x400) | ((t1 >> 18) & 0x300)|((t1 & 0xff00) >> 8)) << 11) |
-		     (((t0 >> 20) & 0x400) | ((t0 >> 16) & 0x300) | (t0 & 0xff));
+void packU96FormatToThreePacket(u32 * out, u8 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t1=U32BIG(((u32*)in)[0]);
+	temp0[0] = t1; temp0[1] = t1 >> 1; temp0[2] = t1>> 2;
+	puckU32ToThree_1(temp0[0]);
+	puckU32ToThree_1(temp0[1]);
+	puckU32ToThree_1(temp0[2]);
+	t1=U32BIG(((u32*)in)[1]);
+	temp1[0] = t1; temp1[1] = t1>>1; temp1[2] = t1 >> 2;
+	puckU32ToThree_1(temp1[0]);
+	puckU32ToThree_1(temp1[1]);
+	puckU32ToThree_1(temp1[2]);
+	t1=U32BIG(((u32*)in)[2]);
+	temp2[0] = t1; temp2[1] =t1 >> 1; temp2[2] = t1>> 2;
+	puckU32ToThree_1(temp2[0]);
+	puckU32ToThree_1(temp2[1]);
+	puckU32ToThree_1(temp2[2]);
+	out[0] = (temp2[1]<<21)	|(temp1[0]<<10)	|temp0[2];
+	out[1] = (temp2[0] << 21) | (temp1[2] << 11) | temp0[1];
+	out[2] = (temp2[2] << 22) | (temp1[1] << 11) | temp0[0];
 }
 }
-void unpackU96FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 } ;\
-u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) |		 \
-((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) |		 \
-((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10) | ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7) |		 \
-((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5) | ((t2 & 0x7f800) >> 3) |		 \
-((t0 & 0x3fc00) >> 10);		 \
-t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1) | ((t2 & 0xc0000000) >> 2) |		 \
-((t0 & 0x60000000) >> 3) | ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6) |		 \
-((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);		 \
-unpuckU32ToThree_3(t[0]);		 \
-unpuckU32ToThree_3(t[1]);		 \
-unpuckU32ToThree_3(t[2]);		 \
-memcpy(out, t, 12 * sizeof(unsigned char));		 \
+void unpackU96FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t[3] = { 0 };
+	u32 t0=in[0] ;
+	u32 t1=in[1] ;
+	u32 t2=in[2] ;
+	temp0[0] = t2 & 0x7ff;
+	temp0[1] = t1 & 0x7ff;
+	temp0[2] = t0 & 0x3ff;
+	temp1[0] = (t0>>10) & 0x7ff;
+	temp1[1] = (t2 >>11 ) & 0x7ff;
+	temp1[2] = (t1 >> 11) & 0x3ff;
+	temp2[0] = t1 >> 21;
+	temp2[1] = t0 >> 21;
+	temp2[2] = t2 >> 22;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;
+	unpuckU32ToThree_1(temp1[0]);
+	unpuckU32ToThree_1(temp1[1]);
+	unpuckU32ToThree_1(temp1[2]);
+	t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;
+	unpuckU32ToThree_1(temp2[0]);
+	unpuckU32ToThree_1(temp2[1]);
+	unpuckU32ToThree_1(temp2[2]);
+	t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;
+	memcpy(out, t, 12 * sizeof(unsigned char));
 }
 }
+
 unsigned char  constant7Format[76] = {
 unsigned char  constant7Format[76] = {
 	/*constant7Format[127]:*/
 	/*constant7Format[127]:*/
 0x01,0x08,0x40,0x02,0x10,0x80,0x05,0x09,0x48,0x42,0x12,0x90,
 0x01,0x08,0x40,0x02,0x10,0x80,0x05,0x09,0x48,0x42,0x12,0x90,

knot/Implementations/crypto_aead/knot192/armcortexm_2/auxFormat.h

@@ -16,25 +16,22 @@ typedef unsigned long long u64;
 
 
 
 
 #define ARR_SIZE(a) (sizeof((a))/sizeof((a[0])))
 #define ARR_SIZE(a) (sizeof((a))/sizeof((a[0])))
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-		u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }
 
 
+
 unsigned char  constant7Format[76];
 unsigned char  constant7Format[76];
 #define P384_ARC_SC1(rci,S2,S3,S4) \
 #define P384_ARC_SC1(rci,S2,S3,S4) \
   do { \
   do { \

knot/Implementations/crypto_aead/knot192/armcortexm_3/auxFormat.h

@@ -22,59 +22,77 @@ unsigned char  constant7Format[76] ;
 	t1 = ~a; t2 = b & t1;t3 = c ^ t2; h = d ^ t3; t5 = b | c; t6 = d ^ t1; g = t5 ^ t6; t8 = b ^ d; t9 = t3 & t6; a = t8 ^ t9; t11 = g & t8; f = t3 ^ t11; \
 	t1 = ~a; t2 = b & t1;t3 = c ^ t2; h = d ^ t3; t5 = b | c; t6 = d ^ t1; g = t5 ^ t6; t8 = b ^ d; t9 = t3 & t6; a = t8 ^ t9; t11 = g & t8; f = t3 ^ t11; \
 }
 }
 
 
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-		u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }
 
 
+
+
 #define packU96FormatToThreePacket( out,  in) {               \
 #define packU96FormatToThreePacket( out,  in) {               \
-		u32  t0 = U32BIG(((u32*)in)[0]), t1 = U32BIG(((u32*)in)[1]), t2 = U32BIG(((u32*)in)[2]);	\
-		puckU32ToThree_3(t0);	\
-		puckU32ToThree_3(t1);	\
-		puckU32ToThree_3(t2);	\
-		out[0] = ((((t2 >> 21) & 0x400) | (((t2 >> 18) & 0x300)) | ((t2 & 0xff00) >> 8)) << 21) |										\
-			     ((((t1 >> 20) & 0x400) | ((t1 >> 16) & 0x300) | (t1 & 0xff))        << 10) |									\
-				 (((t0 >> 20) & 0x300) | ((t0 & 0xff0000) >> 16));								\
-		out[1] = ((((t2 >> 20) & 0x400) | ((t2 >> 16) & 0x300)   | (t2 & 0xff))          << 21) |										\
-			     ((((t1 >> 20) & 0x300) | ((t1 & 0xff0000) >> 16))                   << 11) |									\
-			     (((t0 >> 21) & 0x400) | ((t0 >> 18) & 0x300) | ((t0 & 0xff00) >> 8));									\
-		out[2] = ((((t2 >> 20) & 0x300) | ((t2 & 0xff0000) >> 16))                       << 22) |										\
-			     ((((t1 >> 21) & 0x400) | ((t1 >> 18) & 0x300)|((t1 & 0xff00) >> 8)) << 11) |									\
-			     (((t0 >> 20) & 0x400) | ((t0 >> 16) & 0x300) | (t0 & 0xff));									\
+		t1=U32BIG(((u32*)in)[0]);			\
+		temp0[0] = t1; temp0[1] = t1 >> 1; temp0[2] = t1>> 2;			\
+		puckU32ToThree_1(temp0[0]);			\
+		puckU32ToThree_1(temp0[1]);			\
+		puckU32ToThree_1(temp0[2]);			\
+		t1=U32BIG(((u32*)in)[1]);			\
+		temp1[0] = t1; temp1[1] = t1>>1; temp1[2] = t1 >> 2;			\
+		puckU32ToThree_1(temp1[0]);			\
+		puckU32ToThree_1(temp1[1]);			\
+		puckU32ToThree_1(temp1[2]);			\
+		t1=U32BIG(((u32*)in)[2]);			\
+		temp2[0] = t1; temp2[1] =t1 >> 1; temp2[2] = t1>> 2;			\
+		puckU32ToThree_1(temp2[0]);			\
+		puckU32ToThree_1(temp2[1]);			\
+		puckU32ToThree_1(temp2[2]);			\
+		out[0] = (temp2[1]<<21)	|(temp1[0]<<10)	|temp0[2];	\
+		out[1] = (temp2[0] << 21) | (temp1[2] << 11) | temp0[1];			\
+		out[2] = (temp2[2] << 22) | (temp1[1] << 11) | temp0[0];			\
 }
 }
 
 
 #define unpackU96FormatToThreePacket( out, in) {\
 #define unpackU96FormatToThreePacket( out, in) {\
-		u32 t[3] = { 0 };\
-	u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-	t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) |		 \
-	((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) |		 \
-	((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-	t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10) | ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7) |		 \
-	((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5) | ((t2 & 0x7f800) >> 3) |		 \
-	((t0 & 0x3fc00) >> 10);		 \
-	t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1) | ((t2 & 0xc0000000) >> 2) |		 \
-	((t0 & 0x60000000) >> 3) | ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6) |		 \
-	((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);		 \
-	unpuckU32ToThree_3(t[0]);		 \
-	unpuckU32ToThree_3(t[1]);		 \
-	unpuckU32ToThree_3(t[2]);		 \
+		t3=in[0] ;			\
+		t1=in[1] ;			\
+		t2=in[2] ;			\
+		temp0[0] = t2 & 0x7ff;			\
+		temp0[1] = t1 & 0x7ff;			\
+		temp0[2] = t3 & 0x3ff;			\
+		temp1[0] = (t3>>10) & 0x7ff;			\
+		temp1[1] = (t2 >>11 ) & 0x7ff;			\
+		temp1[2] = (t1 >> 11) & 0x3ff;			\
+		temp2[0] = t1 >> 21;			\
+		temp2[1] = t3 >> 21;			\
+		temp2[2] = t2 >> 22;			\
+		unpuckU32ToThree_1(temp0[0]);			\
+		unpuckU32ToThree_1(temp0[1]);			\
+		unpuckU32ToThree_1(temp0[2]);			\
+		t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;			\
+		unpuckU32ToThree_1(temp1[0]);			\
+		unpuckU32ToThree_1(temp1[1]);			\
+		unpuckU32ToThree_1(temp1[2]);			\
+		t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;			\
+		unpuckU32ToThree_1(temp2[0]);			\
+		unpuckU32ToThree_1(temp2[1]);			\
+		unpuckU32ToThree_1(temp2[2]);			\
+		t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;			\
 		memcpy(out, t, 12 * sizeof(unsigned char));			\
 		memcpy(out, t, 12 * sizeof(unsigned char));			\
 }
 }
 
 
+
+
+
+
+
 #define U96_BIT_LOTR32_8(t0,t1,t2,t3,t4,t5){\
 #define U96_BIT_LOTR32_8(t0,t1,t2,t3,t4,t5){\
 t3= LOTR32(t2, 2);\
 t3= LOTR32(t2, 2);\
 t4 =LOTR32(t0, 3);\
 t4 =LOTR32(t0, 3);\

knot/Implementations/crypto_aead/knot192/armcortexm_3/encrypt.c

@@ -34,6 +34,10 @@ int crypto_aead_encrypt(unsigned char *c, unsigned long long *clen,
 	u8 tempU8[48] = { 0 };
 	u8 tempU8[48] = { 0 };
 	u32 s_temp[12] = { 0 };
 	u32 s_temp[12] = { 0 };
 
 
+	u32 t[3] = { 0 };
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
 	u32 t1, t2, t3, t5, t6, t8, t9, t11;
 	u32 t1, t2, t3, t5, t6, t8, t9, t11;
 	*clen = mlen + CRYPTO_ABYTES;
 	*clen = mlen + CRYPTO_ABYTES;
 	// initialization
 	// initialization
@@ -106,6 +110,10 @@ int crypto_aead_decrypt(unsigned char *m, unsigned long long *mlen,
 	u8 tempU8[48] = { 0 };
 	u8 tempU8[48] = { 0 };
 	u32 s_temp[12] = { 0 };
 	u32 s_temp[12] = { 0 };
 	u32 t1, t2, t3, t5, t6, t8, t9, t11;
 	u32 t1, t2, t3, t5, t6, t8, t9, t11;
+	u32 t[3] = { 0 };
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
 	*mlen = clen - CRYPTO_ABYTES;
 	*mlen = clen - CRYPTO_ABYTES;
 	if (clen < CRYPTO_ABYTES)
 	if (clen < CRYPTO_ABYTES)
 		return -1;
 		return -1;

knot/Implementations/crypto_aead/knot192/armcortexm_4/auxFormat.c

@@ -17,39 +17,61 @@ void P384(unsigned int *s, unsigned char *round, unsigned char lunNum) {
 	}
 	}
 }
 }
 
 
-void packU96FormatToThreePacket(u32 *out, u8 *in) {
-	u32 t0 = U32BIG(((u32*)in)[0]), t1 = U32BIG(((u32*)in)[1]), t2 = U32BIG(((u32*)in)[2]);	\
-	puckU32ToThree_3(t0);	\
-	puckU32ToThree_3(t1);	\
-	puckU32ToThree_3(t2);	\
-	out[0] = ((((t2 >> 21) & 0x400) | (((t2 >> 18) & 0x300)) | ((t2 & 0xff00) >> 8)) << 21) |										\
-		     ((((t1 >> 20) & 0x400) | ((t1 >> 16) & 0x300) | (t1 & 0xff))        << 10) |									\
-			 (((t0 >> 20) & 0x300) | ((t0 & 0xff0000) >> 16));								\
-	out[1] = ((((t2 >> 20) & 0x400) | ((t2 >> 16) & 0x300)   | (t2 & 0xff))          << 21) |										\
-		     ((((t1 >> 20) & 0x300) | ((t1 & 0xff0000) >> 16))                   << 11) |									\
-		     (((t0 >> 21) & 0x400) | ((t0 >> 18) & 0x300) | ((t0 & 0xff00) >> 8));									\
-	out[2] = ((((t2 >> 20) & 0x300) | ((t2 & 0xff0000) >> 16))                       << 22) |										\
-		     ((((t1 >> 21) & 0x400) | ((t1 >> 18) & 0x300)|((t1 & 0xff00) >> 8)) << 11) |									\
-		     (((t0 >> 20) & 0x400) | ((t0 >> 16) & 0x300) | (t0 & 0xff));									\
-
+void packU96FormatToThreePacket(u32 * out, u8 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t1=U32BIG(((u32*)in)[0]);
+	temp0[0] = t1; temp0[1] = t1 >> 1; temp0[2] = t1>> 2;
+	puckU32ToThree_1(temp0[0]);
+	puckU32ToThree_1(temp0[1]);
+	puckU32ToThree_1(temp0[2]);
+	t1=U32BIG(((u32*)in)[1]);
+	temp1[0] = t1; temp1[1] = t1>>1; temp1[2] = t1 >> 2;
+	puckU32ToThree_1(temp1[0]);
+	puckU32ToThree_1(temp1[1]);
+	puckU32ToThree_1(temp1[2]);
+	t1=U32BIG(((u32*)in)[2]);
+	temp2[0] = t1; temp2[1] =t1 >> 1; temp2[2] = t1>> 2;
+	puckU32ToThree_1(temp2[0]);
+	puckU32ToThree_1(temp2[1]);
+	puckU32ToThree_1(temp2[2]);
+	out[0] = (temp2[1]<<21)	|(temp1[0]<<10)	|temp0[2];
+	out[1] = (temp2[0] << 21) | (temp1[2] << 11) | temp0[1];
+	out[2] = (temp2[2] << 22) | (temp1[1] << 11) | temp0[0];
 }
 }
-void unpackU96FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 } ;\
-u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) |		 \
-((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) |		 \
-((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10) | ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7) |		 \
-((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5) | ((t2 & 0x7f800) >> 3) |		 \
-((t0 & 0x3fc00) >> 10);		 \
-t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1) | ((t2 & 0xc0000000) >> 2) |		 \
-((t0 & 0x60000000) >> 3) | ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6) |		 \
-((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);		 \
-unpuckU32ToThree_3(t[0]);		 \
-unpuckU32ToThree_3(t[1]);		 \
-unpuckU32ToThree_3(t[2]);		 \
-memcpy(out, t, 12 * sizeof(unsigned char));		 \
+void unpackU96FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t[3] = { 0 };
+	u32 t0=in[0] ;
+	u32 t1=in[1] ;
+	u32 t2=in[2] ;
+	temp0[0] = t2 & 0x7ff;
+	temp0[1] = t1 & 0x7ff;
+	temp0[2] = t0 & 0x3ff;
+	temp1[0] = (t0>>10) & 0x7ff;
+	temp1[1] = (t2 >>11 ) & 0x7ff;
+	temp1[2] = (t1 >> 11) & 0x3ff;
+	temp2[0] = t1 >> 21;
+	temp2[1] = t0 >> 21;
+	temp2[2] = t2 >> 22;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;
+	unpuckU32ToThree_1(temp1[0]);
+	unpuckU32ToThree_1(temp1[1]);
+	unpuckU32ToThree_1(temp1[2]);
+	t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;
+	unpuckU32ToThree_1(temp2[0]);
+	unpuckU32ToThree_1(temp2[1]);
+	unpuckU32ToThree_1(temp2[2]);
+	t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;
+	memcpy(out, t, 12 * sizeof(unsigned char));
 }
 }
+
 unsigned char  constant7Format[76] = {
 unsigned char  constant7Format[76] = {
 	/*constant7Format[127]:*/
 	/*constant7Format[127]:*/
 0x01,0x08,0x40,0x02,0x10,0x80,0x05,0x09,0x48,0x42,0x12,0x90,
 0x01,0x08,0x40,0x02,0x10,0x80,0x05,0x09,0x48,0x42,0x12,0x90,

knot/Implementations/crypto_aead/knot192/armcortexm_4/auxFormat.h

@@ -38,22 +38,17 @@ t5 = LOTR32(t0, 19); \
 
 
 unsigned char  constant7Format[76];
 unsigned char  constant7Format[76];
 
 
-
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }

knot/Implementations/crypto_aead/knot192/armcortexm_5/auxFormat.c

@@ -29,36 +29,57 @@ void ROUND384_Three(unsigned int *s, unsigned char *c, int lunnum) {
 	}
 	}
 }
 }
 
 
-void packU96FormatToThreePacket(u32 *out, u8 *in) {
-	u32  t0 = U32BIG(((u32*)in)[0]), t1 = U32BIG(((u32*)in)[1]), t2 = U32BIG(((u32*)in)[2]);	\
-	puckU32ToThree_3(t0);	\
-	puckU32ToThree_3(t1);	\
-	puckU32ToThree_3(t2);	\
-	out[0] = ((((t2 >> 21) & 0x400) | (((t2 >> 18) & 0x300)) | ((t2 & 0xff00) >> 8)) << 21) |										\
-		     ((((t1 >> 20) & 0x400) | ((t1 >> 16) & 0x300) | (t1 & 0xff))        << 10) |									\
-			 (((t0 >> 20) & 0x300) | ((t0 & 0xff0000) >> 16));								\
-	out[1] = ((((t2 >> 20) & 0x400) | ((t2 >> 16) & 0x300)   | (t2 & 0xff))          << 21) |										\
-		     ((((t1 >> 20) & 0x300) | ((t1 & 0xff0000) >> 16))                   << 11) |									\
-		     (((t0 >> 21) & 0x400) | ((t0 >> 18) & 0x300) | ((t0 & 0xff00) >> 8));									\
-	out[2] = ((((t2 >> 20) & 0x300) | ((t2 & 0xff0000) >> 16))                       << 22) |										\
-		     ((((t1 >> 21) & 0x400) | ((t1 >> 18) & 0x300)|((t1 & 0xff00) >> 8)) << 11) |									\
-		     (((t0 >> 20) & 0x400) | ((t0 >> 16) & 0x300) | (t0 & 0xff));									\
-
+void packU96FormatToThreePacket(u32 * out, u8 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t1=U32BIG(((u32*)in)[0]);
+	temp0[0] = t1; temp0[1] = t1 >> 1; temp0[2] = t1>> 2;
+	puckU32ToThree_1(temp0[0]);
+	puckU32ToThree_1(temp0[1]);
+	puckU32ToThree_1(temp0[2]);
+	t1=U32BIG(((u32*)in)[1]);
+	temp1[0] = t1; temp1[1] = t1>>1; temp1[2] = t1 >> 2;
+	puckU32ToThree_1(temp1[0]);
+	puckU32ToThree_1(temp1[1]);
+	puckU32ToThree_1(temp1[2]);
+	t1=U32BIG(((u32*)in)[2]);
+	temp2[0] = t1; temp2[1] =t1 >> 1; temp2[2] = t1>> 2;
+	puckU32ToThree_1(temp2[0]);
+	puckU32ToThree_1(temp2[1]);
+	puckU32ToThree_1(temp2[2]);
+	out[0] = (temp2[1]<<21)	|(temp1[0]<<10)	|temp0[2];
+	out[1] = (temp2[0] << 21) | (temp1[2] << 11) | temp0[1];
+	out[2] = (temp2[2] << 22) | (temp1[1] << 11) | temp0[0];
 }
 }
-void unpackU96FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 },r0 ;\
-u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) |		 \
-((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) |		 \
-((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10) | ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7) |		 \
-((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5) | ((t2 & 0x7f800) >> 3) |		 \
-((t0 & 0x3fc00) >> 10);		 \
-t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1) | ((t2 & 0xc0000000) >> 2) |		 \
-((t0 & 0x60000000) >> 3) | ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6) |		 \
-((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);		 \
-unpuckU32ToThree_3(t[0]);		 \
-unpuckU32ToThree_3(t[1]);		 \
-unpuckU32ToThree_3(t[2]);		 \
-memcpy(out, t, 12 * sizeof(unsigned char));		 \
+void unpackU96FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t[3] = { 0 };
+	u32 t0=in[0] ;
+	u32 t1=in[1] ;
+	u32 t2=in[2] ;
+	temp0[0] = t2 & 0x7ff;
+	temp0[1] = t1 & 0x7ff;
+	temp0[2] = t0 & 0x3ff;
+	temp1[0] = (t0>>10) & 0x7ff;
+	temp1[1] = (t2 >>11 ) & 0x7ff;
+	temp1[2] = (t1 >> 11) & 0x3ff;
+	temp2[0] = t1 >> 21;
+	temp2[1] = t0 >> 21;
+	temp2[2] = t2 >> 22;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;
+	unpuckU32ToThree_1(temp1[0]);
+	unpuckU32ToThree_1(temp1[1]);
+	unpuckU32ToThree_1(temp1[2]);
+	t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;
+	unpuckU32ToThree_1(temp2[0]);
+	unpuckU32ToThree_1(temp2[1]);
+	unpuckU32ToThree_1(temp2[2]);
+	t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;
+	memcpy(out, t, 12 * sizeof(unsigned char));
 }
 }

knot/Implementations/crypto_aead/knot192/armcortexm_5/auxFormat.h

@@ -12,22 +12,19 @@ typedef unsigned short u16;
 typedef unsigned int u32;
 typedef unsigned int u32;
 typedef unsigned long long u64;
 typedef unsigned long long u64;
 
 
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }
 void ROUND384_Three(unsigned int *s, unsigned char  *c,int lunnum);
 void ROUND384_Three(unsigned int *s, unsigned char  *c,int lunnum);
 #define ARC(rci) \
 #define ARC(rci) \

knot/Implementations/crypto_aead/knot192/armcortexm_6/auxFormat.c

@@ -28,36 +28,57 @@ void ROUND384_Three(unsigned int *s, unsigned char *c, int lunnum) {
 		t += 3;
 		t += 3;
 	}
 	}
 }
 }
-void packU96FormatToThreePacket(u32 *out, u8 *in) {
-	u32 t0 = U32BIG(((u32*)in)[0]), t1 = U32BIG(((u32*)in)[1]), t2 = U32BIG(((u32*)in)[2]);	\
-	puckU32ToThree_3(t0);	\
-	puckU32ToThree_3(t1);	\
-	puckU32ToThree_3(t2);	\
-	out[0] = ((((t2 >> 21) & 0x400) | (((t2 >> 18) & 0x300)) | ((t2 & 0xff00) >> 8)) << 21) |										\
-		     ((((t1 >> 20) & 0x400) | ((t1 >> 16) & 0x300) | (t1 & 0xff))        << 10) |									\
-			 (((t0 >> 20) & 0x300) | ((t0 & 0xff0000) >> 16));								\
-	out[1] = ((((t2 >> 20) & 0x400) | ((t2 >> 16) & 0x300)   | (t2 & 0xff))          << 21) |										\
-		     ((((t1 >> 20) & 0x300) | ((t1 & 0xff0000) >> 16))                   << 11) |									\
-		     (((t0 >> 21) & 0x400) | ((t0 >> 18) & 0x300) | ((t0 & 0xff00) >> 8));									\
-	out[2] = ((((t2 >> 20) & 0x300) | ((t2 & 0xff0000) >> 16))                       << 22) |										\
-		     ((((t1 >> 21) & 0x400) | ((t1 >> 18) & 0x300)|((t1 & 0xff00) >> 8)) << 11) |									\
-		     (((t0 >> 20) & 0x400) | ((t0 >> 16) & 0x300) | (t0 & 0xff));									\
-
+void packU96FormatToThreePacket(u32 * out, u8 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t1=U32BIG(((u32*)in)[0]);
+	temp0[0] = t1; temp0[1] = t1 >> 1; temp0[2] = t1>> 2;
+	puckU32ToThree_1(temp0[0]);
+	puckU32ToThree_1(temp0[1]);
+	puckU32ToThree_1(temp0[2]);
+	t1=U32BIG(((u32*)in)[1]);
+	temp1[0] = t1; temp1[1] = t1>>1; temp1[2] = t1 >> 2;
+	puckU32ToThree_1(temp1[0]);
+	puckU32ToThree_1(temp1[1]);
+	puckU32ToThree_1(temp1[2]);
+	t1=U32BIG(((u32*)in)[2]);
+	temp2[0] = t1; temp2[1] =t1 >> 1; temp2[2] = t1>> 2;
+	puckU32ToThree_1(temp2[0]);
+	puckU32ToThree_1(temp2[1]);
+	puckU32ToThree_1(temp2[2]);
+	out[0] = (temp2[1]<<21)	|(temp1[0]<<10)	|temp0[2];
+	out[1] = (temp2[0] << 21) | (temp1[2] << 11) | temp0[1];
+	out[2] = (temp2[2] << 22) | (temp1[1] << 11) | temp0[0];
 }
 }
-void unpackU96FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 };\
-u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) |		 \
-((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) |		 \
-((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10) | ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7) |		 \
-((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5) | ((t2 & 0x7f800) >> 3) |		 \
-((t0 & 0x3fc00) >> 10);		 \
-t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1) | ((t2 & 0xc0000000) >> 2) |		 \
-((t0 & 0x60000000) >> 3) | ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6) |		 \
-((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);		 \
-unpuckU32ToThree_3(t[0]);		 \
-unpuckU32ToThree_3(t[1]);		 \
-unpuckU32ToThree_3(t[2]);		 \
-memcpy(out, t, 12 * sizeof(unsigned char));		 \
+void unpackU96FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t[3] = { 0 };
+	u32 t0=in[0] ;
+	u32 t1=in[1] ;
+	u32 t2=in[2] ;
+	temp0[0] = t2 & 0x7ff;
+	temp0[1] = t1 & 0x7ff;
+	temp0[2] = t0 & 0x3ff;
+	temp1[0] = (t0>>10) & 0x7ff;
+	temp1[1] = (t2 >>11 ) & 0x7ff;
+	temp1[2] = (t1 >> 11) & 0x3ff;
+	temp2[0] = t1 >> 21;
+	temp2[1] = t0 >> 21;
+	temp2[2] = t2 >> 22;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;
+	unpuckU32ToThree_1(temp1[0]);
+	unpuckU32ToThree_1(temp1[1]);
+	unpuckU32ToThree_1(temp1[2]);
+	t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;
+	unpuckU32ToThree_1(temp2[0]);
+	unpuckU32ToThree_1(temp2[1]);
+	unpuckU32ToThree_1(temp2[2]);
+	t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;
+	memcpy(out, t, 12 * sizeof(unsigned char));
 }
 }

knot/Implementations/crypto_aead/knot192/armcortexm_6/auxFormat.h

@@ -14,26 +14,20 @@ typedef unsigned long long u64;
 
 
 void ROUND384_Three(unsigned int *s, unsigned char  *c,int lunnum);
 void ROUND384_Three(unsigned int *s, unsigned char  *c,int lunnum);
 
 
-
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-		u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }
-
 #define ARC(rci) \
 #define ARC(rci) \
   do { \
   do { \
     __asm__ __volatile__ ( \
     __asm__ __volatile__ ( \

knot/Implementations/crypto_hash/knot256v2/armcortexm_2/auxFormat.c

 #include"auxFormat.h"
 #include"auxFormat.h"
 
 
-void packU32FormatToThreePacket(u32 *out, u8 *in) {
-	u32 t0 = U32BIG(((u32* )in)[0]);
-	puckU32ToThree_3(t0);
-	out[0] = ((((t0 & 0x30000000) >> 20) | ((t0 & 0xff0000) >> 16)));
-	out[1] = (((t0 & 0x80000000) >> 21) | ((t0 & 0xc000000) >> 18)
-			| ((t0 & 0xff00) >> 8));
-	out[2] = (((t0 & 0x40000000) >> 20) | ((t0 & 0x03000000) >> 16)
-			| (t0 & 0x000000ff));
+void packU32FormatToThreePacket(u32 * out, u8 * in) {
+	u32 t2 = U32BIG(((u32*)in)[0]);
+	out[2] = t2; out[1] = t2 >> 1; out[0] = t2 >> 2;//temp2[0]  0;temp2[1]  1;temp2[2]  2;
+	puckU32ToThree_1(out[0]);
+	puckU32ToThree_1(out[1]);
+	puckU32ToThree_1(out[2]);
 }
 }
-void unpackU32FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 };\
-	u32 t0 = in[0], t1 = in[1], t2 = in[2];
-	t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) | ((t0 & 0x300) << 20)
-			| ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) | ((t0 & 0xff) << 16)
-			| ((t1 & 0xff) << 8) | (t2 & 0xff);
-	unpuckU32ToThree_3(t[0]);
-	*(u32*) (out) = t[0];
+void unpackU32FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	temp0[0] = in[0] & 0x3ff;
+	temp0[1] = in[1] & 0x7ff;
+	temp0[2] = in[2] & 0x7ff;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	*(u32*)(out) = U32BIG(temp0[0]<<2 | temp0[1] << 1 | temp0[2]);
 }
 }
-void packU96FormatToThreePacket(u32 *out, u8 *in) {
-	u32 t0 = U32BIG(((u32* )in)[0]), t1 = U32BIG(((u32* )in)[1]), t2 = U32BIG(
-			((u32* )in)[2]);
-	puckU32ToThree_3(t0);
-	puckU32ToThree_3(t1);
-	puckU32ToThree_3(t2);
-	out[0] = (((t2 & 0x80000000)) | ((t2 & 0xc000000) << 3)
-			| ((t2 & 0xff00) << 13))
-			| (((t1 & 0x40000000) >> 10) | ((t1 & 0x03000000) >> 6)
-					| ((t1 & 0x000000ff) << 10))
-			| ((((t0 & 0x30000000) >> 20) | ((t0 & 0xff0000) >> 16)));
-	out[1] = (((t2 & 0x40000000) << 1) | ((t2 & 0x03000000) << 5)
-			| ((t2 & 0x000000ff) << 21))
-			| ((((t1 & 0x30000000) >> 9) | ((t1 & 0xff0000) >> 5))
-					| (((t0 & 0x80000000) >> 21) | ((t0 & 0xc000000) >> 18)
-							| ((t0 & 0xff00) >> 8)));
-	out[2] = ((((t2 & 0x30000000) << 2) | ((t2 & 0xff0000) << 6))
-			| (((t1 & 0x80000000) >> 10) | ((t1 & 0xc000000) >> 7)
-					| ((t1 & 0xff00) << 3))
-			| (((t0 & 0x40000000) >> 20) | ((t0 & 0x03000000) >> 16)
-					| (t0 & 0x000000ff)));
+
+void packU96FormatToThreePacket(u32 * out, u8 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t1=U32BIG(((u32*)in)[0]);
+	temp0[0] = t1; temp0[1] = t1 >> 1; temp0[2] = t1>> 2;
+	puckU32ToThree_1(temp0[0]);
+	puckU32ToThree_1(temp0[1]);
+	puckU32ToThree_1(temp0[2]);
+	t1=U32BIG(((u32*)in)[1]);
+	temp1[0] = t1; temp1[1] = t1>>1; temp1[2] = t1 >> 2;
+	puckU32ToThree_1(temp1[0]);
+	puckU32ToThree_1(temp1[1]);
+	puckU32ToThree_1(temp1[2]);
+	t1=U32BIG(((u32*)in)[2]);
+	temp2[0] = t1; temp2[1] =t1 >> 1; temp2[2] = t1>> 2;
+	puckU32ToThree_1(temp2[0]);
+	puckU32ToThree_1(temp2[1]);
+	puckU32ToThree_1(temp2[2]);
+	out[0] = (temp2[1]<<21)	|(temp1[0]<<10)	|temp0[2];
+	out[1] = (temp2[0] << 21) | (temp1[2] << 11) | temp0[1];
+	out[2] = (temp2[2] << 22) | (temp1[1] << 11) | temp0[0];
 }
 }
-void unpackU96FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 };\
-	u32 t0 = in[0], t1 = in[1], t2 = in[2];
-	t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) | ((t0 & 0x300) << 20)
-			| ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) | ((t0 & 0xff) << 16)
-			| ((t1 & 0xff) << 8) | (t2 & 0xff);
-	t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10)
-			| ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7)
-			| ((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5)
-			| ((t2 & 0x7f800) >> 3) | ((t0 & 0x3fc00) >> 10);
-	t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1)
-			| ((t2 & 0xc0000000) >> 2) | ((t0 & 0x60000000) >> 3)
-			| ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6)
-			| ((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);
-	unpuckU32ToThree_3(t[0]);
-	unpuckU32ToThree_3(t[1]);
-	unpuckU32ToThree_3(t[2]);
+void unpackU96FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t[3] = { 0 };
+	u32 t0=in[0] ;
+	u32 t1=in[1] ;
+	u32 t2=in[2] ;
+	temp0[0] = t2 & 0x7ff;
+	temp0[1] = t1 & 0x7ff;
+	temp0[2] = t0 & 0x3ff;
+	temp1[0] = (t0>>10) & 0x7ff;
+	temp1[1] = (t2 >>11 ) & 0x7ff;
+	temp1[2] = (t1 >> 11) & 0x3ff;
+	temp2[0] = t1 >> 21;
+	temp2[1] = t0 >> 21;
+	temp2[2] = t2 >> 22;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;
+	unpuckU32ToThree_1(temp1[0]);
+	unpuckU32ToThree_1(temp1[1]);
+	unpuckU32ToThree_1(temp1[2]);
+	t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;
+	unpuckU32ToThree_1(temp2[0]);
+	unpuckU32ToThree_1(temp2[1]);
+	unpuckU32ToThree_1(temp2[2]);
+	t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;
 	memcpy(out, t, 12 * sizeof(unsigned char));
 	memcpy(out, t, 12 * sizeof(unsigned char));
 }
 }
 
 
+
 unsigned char constant7Format[80] = {
 unsigned char constant7Format[80] = {
 /*constant7Format[127]: 12*6=72*/
 /*constant7Format[127]: 12*6=72*/
 0x01, 0x08, 0x40, 0x02, 0x10, 0x80, 0x05, 0x09, 0x48, 0x42, 0x12, 0x90, 0x85,
 0x01, 0x08, 0x40, 0x02, 0x10, 0x80, 0x05, 0x09, 0x48, 0x42, 0x12, 0x90, 0x85,

knot/Implementations/crypto_hash/knot256v2/armcortexm_2/auxFormat.h

@@ -9,23 +9,19 @@
 typedef unsigned char u8;
 typedef unsigned char u8;
 typedef unsigned int u32;
 typedef unsigned int u32;
 typedef unsigned long long u64;
 typedef unsigned long long u64;
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-		u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }
 unsigned char  constant7Format[80];
 unsigned char  constant7Format[80];
 
 

knot/Implementations/crypto_hash/knot256v2/armcortexm_2/hash.c

@@ -12,6 +12,10 @@ int crypto_hash(unsigned char *out, const unsigned char *in,
 	u32 s[12] = { 0 };
 	u32 s[12] = { 0 };
 	u32 dataFormat[6] = { 0 };
 	u32 dataFormat[6] = { 0 };
 	u8 tempData[24] = { 0 };
 	u8 tempData[24] = { 0 };
+	u32 t[3] = { 0 };
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
 	// initialization
 	// initialization
 	s[9] = 0x80000000;
 	s[9] = 0x80000000;
 	//absorb
 	//absorb

knot/Implementations/crypto_hash/knot256v2/armcortexm_3/auxFormat.h

@@ -41,70 +41,84 @@ U96_BIT_LOTR32_8(s_temp[6], s_temp [7], s_temp[ 8], s[6],  s[7], s[8]);\
 U96_BIT_LOTR32_55(s_temp[9], s_temp[10], s_temp[11], s[9], s[10], s[11]);\
 U96_BIT_LOTR32_55(s_temp[9], s_temp[10], s_temp[11], s[9], s[10], s[11]);\
 }
 }
 
 
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-		u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }
+
 #define packU32FormatToThreePacket( out,  in) {\
 #define packU32FormatToThreePacket( out,  in) {\
-u32  t0 = U32BIG(((u32*)in)[0]);	\
-puckU32ToThree_3(t0);	\
-out[0] = ((((t0 & 0x30000000) >> 20) | ((t0 & 0xff0000) >> 16)));	\
-out[1] = (((t0 & 0x80000000) >> 21) | ((t0 & 0xc000000) >> 18) | ((t0 & 0xff00) >> 8));	\
-out[2] = (((t0 & 0x40000000) >> 20) | ((t0 & 0x03000000) >> 16) | (t0 & 0x000000ff));	\
+		u32 t2 = U32BIG(((u32*)in)[0]);			\
+		out[2] = t2; out[1] = t2 >> 1; out[0] = t2 >> 2;	\
+		puckU32ToThree_1(out[0]);			\
+		puckU32ToThree_1(out[1]);			\
+		puckU32ToThree_1(out[2]);			\
 }
 }
 #define unpackU32FormatToThreePacket(out, in) {\
 #define unpackU32FormatToThreePacket(out, in) {\
-u32 t[3] = { 0 }  ;\
-u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) |		 \
-((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) |		 \
-((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-unpuckU32ToThree_3(t[0]);		 \
-*(u32*)(out) = t[0];	\
+		u32 temp0[3] = { 0 };			\
+		temp0[0] = in[0] & 0x3ff;			\
+		temp0[1] = in[1] & 0x7ff;			\
+		temp0[2] = in[2] & 0x7ff;			\
+		unpuckU32ToThree_1(temp0[0]);			\
+		unpuckU32ToThree_1(temp0[1]);			\
+		unpuckU32ToThree_1(temp0[2]);			\
+		*(u32*)(out) = U32BIG(temp0[0]<<2 | temp0[1] << 1 | temp0[2]);			\
 }
 }
 #define packU96FormatToThreePacket( out,  in) {               \
 #define packU96FormatToThreePacket( out,  in) {               \
-	u32  t0 = U32BIG(((u32*)in)[0]), t1 = U32BIG(((u32*)in)[1]), t2 = U32BIG(((u32*)in)[2]);	\
-	puckU32ToThree_3(t0);	\
-	puckU32ToThree_3(t1);	\
-	puckU32ToThree_3(t2);	\
-	out[0] = (((t2 & 0x80000000)) | ((t2 & 0xc000000) << 3) | ((t2 & 0xff00) << 13)) |	\
-	(((t1 & 0x40000000) >> 10) | ((t1 & 0x03000000) >> 6) | ((t1 & 0x000000ff) << 10)) |	\
-	((((t0 & 0x30000000) >> 20) | ((t0 & 0xff0000) >> 16)));	\
-	out[1] = (((t2 & 0x40000000) << 1) | ((t2 & 0x03000000) << 5) | ((t2 & 0x000000ff) << 21)) |	\
-	((((t1 & 0x30000000) >> 9) | ((t1 & 0xff0000) >> 5)) |	\
-	(((t0 & 0x80000000) >> 21) | ((t0 & 0xc000000) >> 18) | ((t0 & 0xff00) >> 8)));	\
-	out[2] = ((((t2 & 0x30000000) << 2) | ((t2 & 0xff0000) << 6)) |	\
-	(((t1 & 0x80000000) >> 10) | ((t1 & 0xc000000) >> 7) | ((t1 & 0xff00) << 3)) |	\
-	(((t0 & 0x40000000) >> 20) | ((t0 & 0x03000000) >> 16) | (t0 & 0x000000ff)));	\
+		t1=U32BIG(((u32*)in)[0]);			\
+		temp0[0] = t1; temp0[1] = t1 >> 1; temp0[2] = t1>> 2;			\
+		puckU32ToThree_1(temp0[0]);			\
+		puckU32ToThree_1(temp0[1]);			\
+		puckU32ToThree_1(temp0[2]);			\
+		t1=U32BIG(((u32*)in)[1]);			\
+		temp1[0] = t1; temp1[1] = t1>>1; temp1[2] = t1 >> 2;			\
+		puckU32ToThree_1(temp1[0]);			\
+		puckU32ToThree_1(temp1[1]);			\
+		puckU32ToThree_1(temp1[2]);			\
+		t1=U32BIG(((u32*)in)[2]);			\
+		temp2[0] = t1; temp2[1] =t1 >> 1; temp2[2] = t1>> 2;			\
+		puckU32ToThree_1(temp2[0]);			\
+		puckU32ToThree_1(temp2[1]);			\
+		puckU32ToThree_1(temp2[2]);			\
+		out[0] = (temp2[1]<<21)	|(temp1[0]<<10)	|temp0[2];	\
+		out[1] = (temp2[0] << 21) | (temp1[2] << 11) | temp0[1];			\
+		out[2] = (temp2[2] << 22) | (temp1[1] << 11) | temp0[0];			\
 }
 }
 
 
 #define unpackU96FormatToThreePacket( out, in) {\
 #define unpackU96FormatToThreePacket( out, in) {\
-	u32 t[3] = { 0 };\
-	u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-	t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) |		 \
-	((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) |		 \
-	((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-	t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10) | ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7) |		 \
-	((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5) | ((t2 & 0x7f800) >> 3) |		 \
-	((t0 & 0x3fc00) >> 10);		 \
-	t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1) | ((t2 & 0xc0000000) >> 2) |		 \
-	((t0 & 0x60000000) >> 3) | ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6) |		 \
-	((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);		 \
-	unpuckU32ToThree_3(t[0]);		 \
-	unpuckU32ToThree_3(t[1]);		 \
-	unpuckU32ToThree_3(t[2]);		 \
+		t3=in[0] ;			\
+		t1=in[1] ;			\
+		t2=in[2] ;			\
+		temp0[0] = t2 & 0x7ff;			\
+		temp0[1] = t1 & 0x7ff;			\
+		temp0[2] = t3 & 0x3ff;			\
+		temp1[0] = (t3>>10) & 0x7ff;			\
+		temp1[1] = (t2 >>11 ) & 0x7ff;			\
+		temp1[2] = (t1 >> 11) & 0x3ff;			\
+		temp2[0] = t1 >> 21;			\
+		temp2[1] = t3 >> 21;			\
+		temp2[2] = t2 >> 22;			\
+		unpuckU32ToThree_1(temp0[0]);			\
+		unpuckU32ToThree_1(temp0[1]);			\
+		unpuckU32ToThree_1(temp0[2]);			\
+		t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;			\
+		unpuckU32ToThree_1(temp1[0]);			\
+		unpuckU32ToThree_1(temp1[1]);			\
+		unpuckU32ToThree_1(temp1[2]);			\
+		t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;			\
+		unpuckU32ToThree_1(temp2[0]);			\
+		unpuckU32ToThree_1(temp2[1]);			\
+		unpuckU32ToThree_1(temp2[2]);			\
+		t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;			\
 		memcpy(out, t, 12 * sizeof(unsigned char));			\
 		memcpy(out, t, 12 * sizeof(unsigned char));			\
 }
 }
+

knot/Implementations/crypto_hash/knot256v2/armcortexm_3/hash.c

@@ -33,6 +33,10 @@ int crypto_hash(unsigned char *out, const unsigned char *in,
 	u8 i, tempData[24] = { 0 };
 	u8 i, tempData[24] = { 0 };
 	u32 s_temp[12] = { 0 };
 	u32 s_temp[12] = { 0 };
 	u32 t1, t2, t3, t5, t6, t8, t9, t11;
 	u32 t1, t2, t3, t5, t6, t8, t9, t11;
+	u32 t[3] = { 0 };
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
 	// initialization
 	// initialization
 	s[9] = 0x80000000;
 	s[9] = 0x80000000;
 	//absorb
 	//absorb

knot/Implementations/crypto_hash/knot256v2/armcortexm_4/auxFormat.c

 #include"auxFormat.h"
 #include"auxFormat.h"
 
 
-void packU32FormatToThreePacket(u32 *out, u8 *in) {
-	u32 t0 = U32BIG(((u32* )in)[0]);
-	puckU32ToThree_3(t0);
-	out[0] = ((((t0 & 0x30000000) >> 20) | ((t0 & 0xff0000) >> 16)));
-	out[1] = (((t0 & 0x80000000) >> 21) | ((t0 & 0xc000000) >> 18)
-			| ((t0 & 0xff00) >> 8));
-	out[2] = (((t0 & 0x40000000) >> 20) | ((t0 & 0x03000000) >> 16)
-			| (t0 & 0x000000ff));
+void packU32FormatToThreePacket(u32 * out, u8 * in) {
+	u32 t2 = U32BIG(((u32*)in)[0]);
+	out[2] = t2; out[1] = t2 >> 1; out[0] = t2 >> 2;//temp2[0]  0;temp2[1]  1;temp2[2]  2;
+	puckU32ToThree_1(out[0]);
+	puckU32ToThree_1(out[1]);
+	puckU32ToThree_1(out[2]);
 }
 }
-void unpackU32FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 };\
-	u32 t0 = in[0], t1 = in[1], t2 = in[2];
-	t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) | ((t0 & 0x300) << 20)
-			| ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) | ((t0 & 0xff) << 16)
-			| ((t1 & 0xff) << 8) | (t2 & 0xff);
-	unpuckU32ToThree_3(t[0]);
-	*(u32*) (out) = t[0];
+void unpackU32FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	temp0[0] = in[0] & 0x3ff;
+	temp0[1] = in[1] & 0x7ff;
+	temp0[2] = in[2] & 0x7ff;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	*(u32*)(out) = U32BIG(temp0[0]<<2 | temp0[1] << 1 | temp0[2]);
 }
 }
-void packU96FormatToThreePacket(u32 *out, u8 *in) {
-	u32 t0 = U32BIG(((u32* )in)[0]), t1 = U32BIG(((u32* )in)[1]), t2 = U32BIG(
-			((u32* )in)[2]);
-	puckU32ToThree_3(t0);
-	puckU32ToThree_3(t1);
-	puckU32ToThree_3(t2);
-	out[0] = (((t2 & 0x80000000)) | ((t2 & 0xc000000) << 3)
-			| ((t2 & 0xff00) << 13))
-			| (((t1 & 0x40000000) >> 10) | ((t1 & 0x03000000) >> 6)
-					| ((t1 & 0x000000ff) << 10))
-			| ((((t0 & 0x30000000) >> 20) | ((t0 & 0xff0000) >> 16)));
-	out[1] = (((t2 & 0x40000000) << 1) | ((t2 & 0x03000000) << 5)
-			| ((t2 & 0x000000ff) << 21))
-			| ((((t1 & 0x30000000) >> 9) | ((t1 & 0xff0000) >> 5))
-					| (((t0 & 0x80000000) >> 21) | ((t0 & 0xc000000) >> 18)
-							| ((t0 & 0xff00) >> 8)));
-	out[2] = ((((t2 & 0x30000000) << 2) | ((t2 & 0xff0000) << 6))
-			| (((t1 & 0x80000000) >> 10) | ((t1 & 0xc000000) >> 7)
-					| ((t1 & 0xff00) << 3))
-			| (((t0 & 0x40000000) >> 20) | ((t0 & 0x03000000) >> 16)
-					| (t0 & 0x000000ff)));
+
+void packU96FormatToThreePacket(u32 * out, u8 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t1=U32BIG(((u32*)in)[0]);
+	temp0[0] = t1; temp0[1] = t1 >> 1; temp0[2] = t1>> 2;
+	puckU32ToThree_1(temp0[0]);
+	puckU32ToThree_1(temp0[1]);
+	puckU32ToThree_1(temp0[2]);
+	t1=U32BIG(((u32*)in)[1]);
+	temp1[0] = t1; temp1[1] = t1>>1; temp1[2] = t1 >> 2;
+	puckU32ToThree_1(temp1[0]);
+	puckU32ToThree_1(temp1[1]);
+	puckU32ToThree_1(temp1[2]);
+	t1=U32BIG(((u32*)in)[2]);
+	temp2[0] = t1; temp2[1] =t1 >> 1; temp2[2] = t1>> 2;
+	puckU32ToThree_1(temp2[0]);
+	puckU32ToThree_1(temp2[1]);
+	puckU32ToThree_1(temp2[2]);
+	out[0] = (temp2[1]<<21)	|(temp1[0]<<10)	|temp0[2];
+	out[1] = (temp2[0] << 21) | (temp1[2] << 11) | temp0[1];
+	out[2] = (temp2[2] << 22) | (temp1[1] << 11) | temp0[0];
 }
 }
-void unpackU96FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 };\
-	u32 t0 = in[0], t1 = in[1], t2 = in[2];
-	t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) | ((t0 & 0x300) << 20)
-			| ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) | ((t0 & 0xff) << 16)
-			| ((t1 & 0xff) << 8) | (t2 & 0xff);
-	t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10)
-			| ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7)
-			| ((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5)
-			| ((t2 & 0x7f800) >> 3) | ((t0 & 0x3fc00) >> 10);
-	t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1)
-			| ((t2 & 0xc0000000) >> 2) | ((t0 & 0x60000000) >> 3)
-			| ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6)
-			| ((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);
-	unpuckU32ToThree_3(t[0]);
-	unpuckU32ToThree_3(t[1]);
-	unpuckU32ToThree_3(t[2]);
+void unpackU96FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t[3] = { 0 };
+	u32 t0=in[0] ;
+	u32 t1=in[1] ;
+	u32 t2=in[2] ;
+	temp0[0] = t2 & 0x7ff;
+	temp0[1] = t1 & 0x7ff;
+	temp0[2] = t0 & 0x3ff;
+	temp1[0] = (t0>>10) & 0x7ff;
+	temp1[1] = (t2 >>11 ) & 0x7ff;
+	temp1[2] = (t1 >> 11) & 0x3ff;
+	temp2[0] = t1 >> 21;
+	temp2[1] = t0 >> 21;
+	temp2[2] = t2 >> 22;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;
+	unpuckU32ToThree_1(temp1[0]);
+	unpuckU32ToThree_1(temp1[1]);
+	unpuckU32ToThree_1(temp1[2]);
+	t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;
+	unpuckU32ToThree_1(temp2[0]);
+	unpuckU32ToThree_1(temp2[1]);
+	unpuckU32ToThree_1(temp2[2]);
+	t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;
 	memcpy(out, t, 12 * sizeof(unsigned char));
 	memcpy(out, t, 12 * sizeof(unsigned char));
 }
 }
+
 unsigned char constant7Format[80] = {
 unsigned char constant7Format[80] = {
 /*constant7Format[127]: 12*6=72*/
 /*constant7Format[127]: 12*6=72*/
 0x01, 0x08, 0x40, 0x02, 0x10, 0x80, 0x05, 0x09, 0x48, 0x42, 0x12, 0x90, 0x85,
 0x01, 0x08, 0x40, 0x02, 0x10, 0x80, 0x05, 0x09, 0x48, 0x42, 0x12, 0x90, 0x85,

knot/Implementations/crypto_hash/knot256v2/armcortexm_4/auxFormat.h

@@ -14,23 +14,19 @@ void unpackU96FormatToThreePacket(u8 * out, u32 * in) ;
 void packU96FormatToThreePacket(u32 * out, u8 * in);
 void packU96FormatToThreePacket(u32 * out, u8 * in);
 void unpackU32FormatToThreePacket(u8 * out, u32 * in);
 void unpackU32FormatToThreePacket(u8 * out, u32 * in);
 void packU32FormatToThreePacket(u32 * out, u8 * in);
 void packU32FormatToThreePacket(u32 * out, u8 * in);
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-		u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }
 unsigned char  constant7Format[80];
 unsigned char  constant7Format[80];
 
 

knot/Implementations/crypto_hash/knot256v2/armcortexm_5/auxFormat.c

 #include"auxFormat.h"
 #include"auxFormat.h"
 
 
-void packU32FormatToThreePacket(u32 *out, u8 *in) {
-	u32 t0 = U32BIG(((u32* )in)[0]);
-	puckU32ToThree_3(t0);
-	out[0] = ((((t0 & 0x30000000) >> 20) | ((t0 & 0xff0000) >> 16)));
-	out[1] = (((t0 & 0x80000000) >> 21) | ((t0 & 0xc000000) >> 18)
-			| ((t0 & 0xff00) >> 8));
-	out[2] = (((t0 & 0x40000000) >> 20) | ((t0 & 0x03000000) >> 16)
-			| (t0 & 0x000000ff));
+
+void packU32FormatToThreePacket(u32 * out, u8 * in) {
+	u32 t2 = U32BIG(((u32*)in)[0]);
+	out[2] = t2; out[1] = t2 >> 1; out[0] = t2 >> 2;//temp2[0]  0;temp2[1]  1;temp2[2]  2;
+	puckU32ToThree_1(out[0]);
+	puckU32ToThree_1(out[1]);
+	puckU32ToThree_1(out[2]);
 }
 }
-void unpackU32FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 };\
-	u32 t0 = in[0], t1 = in[1], t2 = in[2];
-	t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) | ((t0 & 0x300) << 20)
-			| ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) | ((t0 & 0xff) << 16)
-			| ((t1 & 0xff) << 8) | (t2 & 0xff);
-	unpuckU32ToThree_3(t[0]);
-	*(u32*) (out) = t[0];
+void unpackU32FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	temp0[0] = in[0] & 0x3ff;
+	temp0[1] = in[1] & 0x7ff;
+	temp0[2] = in[2] & 0x7ff;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	*(u32*)(out) = U32BIG(temp0[0]<<2 | temp0[1] << 1 | temp0[2]);
 }
 }
-void packU96FormatToThreePacket(u32 *out, u8 *in) {
-	u32 t0 = U32BIG(((u32* )in)[0]), t1 = U32BIG(((u32* )in)[1]), t2 = U32BIG(
-			((u32* )in)[2]);
-	puckU32ToThree_3(t0);
-	puckU32ToThree_3(t1);
-	puckU32ToThree_3(t2);
-	out[0] = (((t2 & 0x80000000)) | ((t2 & 0xc000000) << 3)
-			| ((t2 & 0xff00) << 13))
-			| (((t1 & 0x40000000) >> 10) | ((t1 & 0x03000000) >> 6)
-					| ((t1 & 0x000000ff) << 10))
-			| ((((t0 & 0x30000000) >> 20) | ((t0 & 0xff0000) >> 16)));
-	out[1] = (((t2 & 0x40000000) << 1) | ((t2 & 0x03000000) << 5)
-			| ((t2 & 0x000000ff) << 21))
-			| ((((t1 & 0x30000000) >> 9) | ((t1 & 0xff0000) >> 5))
-					| (((t0 & 0x80000000) >> 21) | ((t0 & 0xc000000) >> 18)
-							| ((t0 & 0xff00) >> 8)));
-	out[2] = ((((t2 & 0x30000000) << 2) | ((t2 & 0xff0000) << 6))
-			| (((t1 & 0x80000000) >> 10) | ((t1 & 0xc000000) >> 7)
-					| ((t1 & 0xff00) << 3))
-			| (((t0 & 0x40000000) >> 20) | ((t0 & 0x03000000) >> 16)
-					| (t0 & 0x000000ff)));
+
+void packU96FormatToThreePacket(u32 * out, u8 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t1=U32BIG(((u32*)in)[0]);
+	temp0[0] = t1; temp0[1] = t1 >> 1; temp0[2] = t1>> 2;
+	puckU32ToThree_1(temp0[0]);
+	puckU32ToThree_1(temp0[1]);
+	puckU32ToThree_1(temp0[2]);
+	t1=U32BIG(((u32*)in)[1]);
+	temp1[0] = t1; temp1[1] = t1>>1; temp1[2] = t1 >> 2;
+	puckU32ToThree_1(temp1[0]);
+	puckU32ToThree_1(temp1[1]);
+	puckU32ToThree_1(temp1[2]);
+	t1=U32BIG(((u32*)in)[2]);
+	temp2[0] = t1; temp2[1] =t1 >> 1; temp2[2] = t1>> 2;
+	puckU32ToThree_1(temp2[0]);
+	puckU32ToThree_1(temp2[1]);
+	puckU32ToThree_1(temp2[2]);
+	out[0] = (temp2[1]<<21)	|(temp1[0]<<10)	|temp0[2];
+	out[1] = (temp2[0] << 21) | (temp1[2] << 11) | temp0[1];
+	out[2] = (temp2[2] << 22) | (temp1[1] << 11) | temp0[0];
 }
 }
-void unpackU96FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 };\
-	u32 t0 = in[0], t1 = in[1], t2 = in[2];
-	t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) | ((t0 & 0x300) << 20)
-			| ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) | ((t0 & 0xff) << 16)
-			| ((t1 & 0xff) << 8) | (t2 & 0xff);
-	t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10)
-			| ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7)
-			| ((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5)
-			| ((t2 & 0x7f800) >> 3) | ((t0 & 0x3fc00) >> 10);
-	t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1)
-			| ((t2 & 0xc0000000) >> 2) | ((t0 & 0x60000000) >> 3)
-			| ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6)
-			| ((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);
-	unpuckU32ToThree_3(t[0]);
-	unpuckU32ToThree_3(t[1]);
-	unpuckU32ToThree_3(t[2]);
+void unpackU96FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t[3] = { 0 };
+	u32 t0=in[0] ;
+	u32 t1=in[1] ;
+	u32 t2=in[2] ;
+	temp0[0] = t2 & 0x7ff;
+	temp0[1] = t1 & 0x7ff;
+	temp0[2] = t0 & 0x3ff;
+	temp1[0] = (t0>>10) & 0x7ff;
+	temp1[1] = (t2 >>11 ) & 0x7ff;
+	temp1[2] = (t1 >> 11) & 0x3ff;
+	temp2[0] = t1 >> 21;
+	temp2[1] = t0 >> 21;
+	temp2[2] = t2 >> 22;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;
+	unpuckU32ToThree_1(temp1[0]);
+	unpuckU32ToThree_1(temp1[1]);
+	unpuckU32ToThree_1(temp1[2]);
+	t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;
+	unpuckU32ToThree_1(temp2[0]);
+	unpuckU32ToThree_1(temp2[1]);
+	unpuckU32ToThree_1(temp2[2]);
+	t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;
 	memcpy(out, t, 12 * sizeof(unsigned char));
 	memcpy(out, t, 12 * sizeof(unsigned char));
 }
 }
+
 void ROUND384_Three(unsigned int *s, unsigned char *c, int lunnum) {
 void ROUND384_Three(unsigned int *s, unsigned char *c, int lunnum) {
 	unsigned int t, t1, t2;
 	unsigned int t, t1, t2;
 	u32 rci;
 	u32 rci;

knot/Implementations/crypto_hash/knot256v2/armcortexm_5/auxFormat.h

@@ -15,23 +15,19 @@ void unpackU96FormatToThreePacket(u8 * out, u32 * in) ;
 void packU96FormatToThreePacket(u32 * out, u8 * in);
 void packU96FormatToThreePacket(u32 * out, u8 * in);
 void unpackU32FormatToThreePacket(u8 * out, u32 * in);
 void unpackU32FormatToThreePacket(u8 * out, u32 * in);
 void packU32FormatToThreePacket(u32 * out, u8 * in);
 void packU32FormatToThreePacket(u32 * out, u8 * in);
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-		u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }
 ////////////constant begin//
 ////////////constant begin//
 unsigned char  constant7Format[80];
 unsigned char  constant7Format[80];

knot/Implementations/crypto_hash/knot256v2/armcortexm_6/auxFormat.c

 #include"auxFormat.h"
 #include"auxFormat.h"
 
 
-void packU32FormatToThreePacket(u32 *out, u8 *in) {
-	u32   t0 = U32BIG(((u32*)in)[0]);	\
-	puckU32ToThree_3(t0);	\
-	out[0] = ((((t0 & 0x30000000) >> 20) | ((t0 & 0xff0000) >> 16)));	\
-out[1] = (((t0 & 0x80000000) >> 21) | ((t0 & 0xc000000) >> 18) | ((t0 & 0xff00) >> 8));	\
-out[2] = (((t0 & 0x40000000) >> 20) | ((t0 & 0x03000000) >> 16) | (t0 & 0x000000ff));	\
+
+void packU32FormatToThreePacket(u32 * out, u8 * in) {
+	u32 t2 = U32BIG(((u32*)in)[0]);
+	out[2] = t2; out[1] = t2 >> 1; out[0] = t2 >> 2;//temp2[0]  0;temp2[1]  1;temp2[2]  2;
+	puckU32ToThree_1(out[0]);
+	puckU32ToThree_1(out[1]);
+	puckU32ToThree_1(out[2]);
 }
 }
-void unpackU32FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 }  ;\
-u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) |		 \
-((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) |		 \
-((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-unpuckU32ToThree_3(t[0]);		 \
-*(u32*)(out) = t[0];	\
+void unpackU32FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	temp0[0] = in[0] & 0x3ff;
+	temp0[1] = in[1] & 0x7ff;
+	temp0[2] = in[2] & 0x7ff;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	*(u32*)(out) = U32BIG(temp0[0]<<2 | temp0[1] << 1 | temp0[2]);
 }
 }
-void packU96FormatToThreePacket(u32 *out, u8 *in) {
-	u32 t0 = U32BIG(((u32*)in)[0]), t1 = U32BIG(((u32*)in)[1]), t2 = U32BIG(((u32*)in)[2]);	\
-	puckU32ToThree_3(t0);	\
-	puckU32ToThree_3(t1);	\
-	puckU32ToThree_3(t2);	\
-	out[0] = (((t2 & 0x80000000)) | ((t2 & 0xc000000) << 3) | ((t2 & 0xff00) << 13)) |	\
-(((t1 & 0x40000000) >> 10) | ((t1 & 0x03000000) >> 6) | ((t1 & 0x000000ff) << 10)) |	\
-((((t0 & 0x30000000) >> 20) | ((t0 & 0xff0000) >> 16)));	\
-out[1] = (((t2 & 0x40000000) << 1) | ((t2 & 0x03000000) << 5) | ((t2 & 0x000000ff) << 21)) |	\
-((((t1 & 0x30000000) >> 9) | ((t1 & 0xff0000) >> 5)) |	\
-(((t0 & 0x80000000) >> 21) | ((t0 & 0xc000000) >> 18) | ((t0 & 0xff00) >> 8)));	\
-out[2] = ((((t2 & 0x30000000) << 2) | ((t2 & 0xff0000) << 6)) |	\
-(((t1 & 0x80000000) >> 10) | ((t1 & 0xc000000) >> 7) | ((t1 & 0xff00) << 3)) |	\
-(((t0 & 0x40000000) >> 20) | ((t0 & 0x03000000) >> 16) | (t0 & 0x000000ff)));	\
+
+void packU96FormatToThreePacket(u32 * out, u8 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t1=U32BIG(((u32*)in)[0]);
+	temp0[0] = t1; temp0[1] = t1 >> 1; temp0[2] = t1>> 2;
+	puckU32ToThree_1(temp0[0]);
+	puckU32ToThree_1(temp0[1]);
+	puckU32ToThree_1(temp0[2]);
+	t1=U32BIG(((u32*)in)[1]);
+	temp1[0] = t1; temp1[1] = t1>>1; temp1[2] = t1 >> 2;
+	puckU32ToThree_1(temp1[0]);
+	puckU32ToThree_1(temp1[1]);
+	puckU32ToThree_1(temp1[2]);
+	t1=U32BIG(((u32*)in)[2]);
+	temp2[0] = t1; temp2[1] =t1 >> 1; temp2[2] = t1>> 2;
+	puckU32ToThree_1(temp2[0]);
+	puckU32ToThree_1(temp2[1]);
+	puckU32ToThree_1(temp2[2]);
+	out[0] = (temp2[1]<<21)	|(temp1[0]<<10)	|temp0[2];
+	out[1] = (temp2[0] << 21) | (temp1[2] << 11) | temp0[1];
+	out[2] = (temp2[2] << 22) | (temp1[1] << 11) | temp0[0];
 }
 }
-void unpackU96FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 };\
-u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) |		 \
-((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) |		 \
-((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10) | ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7) |		 \
-((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5) | ((t2 & 0x7f800) >> 3) |		 \
-((t0 & 0x3fc00) >> 10);		 \
-t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1) | ((t2 & 0xc0000000) >> 2) |		 \
-((t0 & 0x60000000) >> 3) | ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6) |		 \
-((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);		 \
-unpuckU32ToThree_3(t[0]);		 \
-unpuckU32ToThree_3(t[1]);		 \
-unpuckU32ToThree_3(t[2]);		 \
-memcpy(out, t, 12 * sizeof(unsigned char));		 \
+void unpackU96FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t[3] = { 0 };
+	u32 t0=in[0] ;
+	u32 t1=in[1] ;
+	u32 t2=in[2] ;
+	temp0[0] = t2 & 0x7ff;
+	temp0[1] = t1 & 0x7ff;
+	temp0[2] = t0 & 0x3ff;
+	temp1[0] = (t0>>10) & 0x7ff;
+	temp1[1] = (t2 >>11 ) & 0x7ff;
+	temp1[2] = (t1 >> 11) & 0x3ff;
+	temp2[0] = t1 >> 21;
+	temp2[1] = t0 >> 21;
+	temp2[2] = t2 >> 22;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;
+	unpuckU32ToThree_1(temp1[0]);
+	unpuckU32ToThree_1(temp1[1]);
+	unpuckU32ToThree_1(temp1[2]);
+	t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;
+	unpuckU32ToThree_1(temp2[0]);
+	unpuckU32ToThree_1(temp2[1]);
+	unpuckU32ToThree_1(temp2[2]);
+	t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;
+	memcpy(out, t, 12 * sizeof(unsigned char));
 }
 }
 void ROUND384_Three(unsigned int *s, unsigned char *c, int lunnum) {
 void ROUND384_Three(unsigned int *s, unsigned char *c, int lunnum) {
 	unsigned int t, t1, t2;
 	unsigned int t, t1, t2;

knot/Implementations/crypto_hash/knot256v2/armcortexm_6/auxFormat.h

@@ -11,24 +11,21 @@ typedef unsigned int u32;
 typedef unsigned long long u64;
 typedef unsigned long long u64;
 
 
 void ROUND384_Three(unsigned int *s, unsigned char  *c,int lunnum) ;
 void ROUND384_Three(unsigned int *s, unsigned char  *c,int lunnum) ;
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-		u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }
+
 ////////////constant begin//
 ////////////constant begin//
 unsigned char  constant7Format[80];
 unsigned char  constant7Format[80];
 #define ARC(rci) \
 #define ARC(rci) \

knot/Implementations/crypto_hash/knot384/armcortexm_2/auxFormat.c

 #include"auxFormat.h"
 #include"auxFormat.h"
-
-void unpackU96FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 }; \
-		u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-		t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) | \
-		((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) | \
-		((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-		t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10) | ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7) | \
-		((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5) | ((t2 & 0x7f800) >> 3) | \
-		((t0 & 0x3fc00) >> 10);		 \
-		t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1) | ((t2 & 0xc0000000) >> 2) | \
-		((t0 & 0x60000000) >> 3) | ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6) | \
-		((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);		 \
-		unpuckU32ToThree_3(t[0]);		 \
-		unpuckU32ToThree_3(t[1]);		 \
-		unpuckU32ToThree_3(t[2]);		 \
-		memcpy(out, t, 12 * sizeof(unsigned char));		 \
+void unpackU96FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t[3] = { 0 };
+	temp0[0] = in[2] & 0x7ff;
+	temp0[1] = in[1] & 0x7ff;
+	temp0[2] = in[0] & 0x3ff;
+	temp1[0] = (in[0]>>10) & 0x7ff;
+	temp1[1] = (in[2] >>11 ) & 0x7ff;
+	temp1[2] = (in[1] >> 11) & 0x3ff;
+	temp2[0] = in[1] >> 21;
+	temp2[1] = in[0] >> 21;
+	temp2[2] = in[2] >> 22;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;
+	unpuckU32ToThree_1(temp1[0]);
+	unpuckU32ToThree_1(temp1[1]);
+	unpuckU32ToThree_1(temp1[2]);
+	t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;
+	unpuckU32ToThree_1(temp2[0]);
+	unpuckU32ToThree_1(temp2[1]);
+	unpuckU32ToThree_1(temp2[2]);
+	t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;
+	memcpy(out, t, 12 * sizeof(unsigned char));
 }
 }
+
 void packU48FormatToThreePacket(u32 * out, u8 * in) {
 void packU48FormatToThreePacket(u32 * out, u8 * in) {
-	u32 t0 = U32BIG(*(u32*)(in)), t1 = (u32)U16BIG(*(u16*)(in + 4));	\
-		puckU32ToThree_3(t0);	\
-		puckU32ToThree_3(t1);	\
-		out[0] =  \
-		(((t1 & 0x40000000) >> 10) | ((t1 & 0x03000000) >> 6) | ((t1 & 0x000000ff) << 10)) | \
-		((((t0 & 0x30000000) >> 20) | ((t0 & 0xff0000) >> 16)));	\
-		out[1] =  \
-		((((t1 & 0x30000000) >> 9) | ((t1 & 0xff0000) >> 5)) | \
-		(((t0 & 0x80000000) >> 21) | ((t0 & 0xc000000) >> 18) | ((t0 & 0xff00) >> 8)));	\
-		out[2] =  \
-		(((t1 & 0x80000000) >> 10) | ((t1 & 0xc000000) >> 7) | ((t1 & 0xff00) << 3)) | \
-			(((t0 & 0x40000000) >> 20) | ((t0 & 0x03000000) >> 16) | (t0 & 0x000000ff));	\
+	u32 t1 = (u32)U16BIG(*(u16*)(in + 4));
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	temp0[0] = U32BIG(((u32*)in)[0]); temp0[1] = U32BIG(((u32*)in)[0]) >> 1; temp0[2] = U32BIG(((u32*)in)[0]) >> 2;
+	puckU32ToThree_1(temp0[0]);
+	puckU32ToThree_1(temp0[1]);
+	puckU32ToThree_1(temp0[2]);
+	temp1[0] = t1; temp1[1] = t1 >> 1; temp1[2] = t1 >> 2;
+	puckU32ToThree_1(temp1[0]);
+	puckU32ToThree_1(temp1[1]);
+	puckU32ToThree_1(temp1[2]);
+	out[0] =  (temp1[0] << 10) | temp0[2];
+	out[1] =  (temp1[2] << 11) | temp0[1];
+	out[2] = (temp1[1] << 11) | temp0[0];
 }
 }
 
 
 unsigned char  constant7Format[104] = {
 unsigned char  constant7Format[104] = {

knot/Implementations/crypto_hash/knot384/armcortexm_2/auxFormat.h

@@ -17,24 +17,21 @@ void packU48FormatToThreePacket(u32 * out, u8 * in) ;
 
 
 void P384(unsigned int *s, unsigned char *round, unsigned char lunNum) ;
 void P384(unsigned int *s, unsigned char *round, unsigned char lunNum) ;
 void unpackU96FormatToThreePacket(u8 * out, u32 * in) ;
 void unpackU96FormatToThreePacket(u8 * out, u32 * in) ;
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-		u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }
+
 #define P384_ARC_SC1(rci,S2,S3,S4) \
 #define P384_ARC_SC1(rci,S2,S3,S4) \
   do { \
   do { \
     __asm__ __volatile__ ( \
     __asm__ __volatile__ ( \

knot/Implementations/crypto_hash/knot384/armcortexm_3/auxFormat.h

@@ -22,55 +22,62 @@ typedef unsigned long long u64;
 {                                                                                                                             \
 {                                                                                                                             \
 	t1 = ~a; t2 = b & t1;t3 = c ^ t2; h = d ^ t3; t5 = b | c; t6 = d ^ t1; g = t5 ^ t6; t8 = b ^ d; t9 = t3 & t6; a = t8 ^ t9; t11 = g & t8; f = t3 ^ t11; \
 	t1 = ~a; t2 = b & t1;t3 = c ^ t2; h = d ^ t3; t5 = b | c; t6 = d ^ t1; g = t5 ^ t6; t8 = b ^ d; t9 = t3 & t6; a = t8 ^ t9; t11 = g & t8; f = t3 ^ t11; \
 }
 }
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-		u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }
 
 
+
 #define unpackU96FormatToThreePacket( out, in) {\
 #define unpackU96FormatToThreePacket( out, in) {\
-		u32 t[3] = { 0 };\
-	u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-	t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) |		 \
-	((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) |		 \
-	((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-	t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10) | ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7) |		 \
-	((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5) | ((t2 & 0x7f800) >> 3) |		 \
-	((t0 & 0x3fc00) >> 10);		 \
-	t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1) | ((t2 & 0xc0000000) >> 2) |		 \
-	((t0 & 0x60000000) >> 3) | ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6) |		 \
-	((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);		 \
-	unpuckU32ToThree_3(t[0]);		 \
-	unpuckU32ToThree_3(t[1]);		 \
-	unpuckU32ToThree_3(t[2]);		 \
+		t3=in[0] ;			\
+		t1=in[1] ;			\
+		t2=in[2] ;			\
+		temp0[0] = t2 & 0x7ff;			\
+		temp0[1] = t1 & 0x7ff;			\
+		temp0[2] = t3 & 0x3ff;			\
+		temp1[0] = (t3>>10) & 0x7ff;			\
+		temp1[1] = (t2 >>11 ) & 0x7ff;			\
+		temp1[2] = (t1 >> 11) & 0x3ff;			\
+		temp2[0] = t1 >> 21;			\
+		temp2[1] = t3 >> 21;			\
+		temp2[2] = t2 >> 22;			\
+		unpuckU32ToThree_1(temp0[0]);			\
+		unpuckU32ToThree_1(temp0[1]);			\
+		unpuckU32ToThree_1(temp0[2]);			\
+		t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;			\
+		unpuckU32ToThree_1(temp1[0]);			\
+		unpuckU32ToThree_1(temp1[1]);			\
+		unpuckU32ToThree_1(temp1[2]);			\
+		t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;			\
+		unpuckU32ToThree_1(temp2[0]);			\
+		unpuckU32ToThree_1(temp2[1]);			\
+		unpuckU32ToThree_1(temp2[2]);			\
+		t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;			\
 		memcpy(out, t, 12 * sizeof(unsigned char));			\
 		memcpy(out, t, 12 * sizeof(unsigned char));			\
 }
 }
 #define packU48FormatToThreePacket(out, in) {\
 #define packU48FormatToThreePacket(out, in) {\
-	u32 t0 = U32BIG(*(u32*)(in)), t1 = (u32)U16BIG(*(u16*)(in + 4));	\
-		puckU32ToThree_3(t0);	\
-		puckU32ToThree_3(t1);	\
-		out[0] =  \
-		(((t1 & 0x40000000) >> 10) | ((t1 & 0x03000000) >> 6) | ((t1 & 0x000000ff) << 10)) | \
-		((((t0 & 0x30000000) >> 20) | ((t0 & 0xff0000) >> 16)));	\
-		out[1] =  \
-		((((t1 & 0x30000000) >> 9) | ((t1 & 0xff0000) >> 5)) | \
-		(((t0 & 0x80000000) >> 21) | ((t0 & 0xc000000) >> 18) | ((t0 & 0xff00) >> 8)));	\
-		out[2] =  \
-		(((t1 & 0x80000000) >> 10) | ((t1 & 0xc000000) >> 7) | ((t1 & 0xff00) << 3)) | \
-			(((t0 & 0x40000000) >> 20) | ((t0 & 0x03000000) >> 16) | (t0 & 0x000000ff));	\
+		t1 = (u32)U16BIG(*(u16*)(in + 4));		\
+		temp0[0] = U32BIG(((u32*)in)[0]); temp0[1] = U32BIG(((u32*)in)[0]) >> 1; temp0[2] = U32BIG(((u32*)in)[0]) >> 2;		\
+		puckU32ToThree_1(temp0[0]);		\
+		puckU32ToThree_1(temp0[1]);		\
+		puckU32ToThree_1(temp0[2]);		\
+		temp1[0] = t1; temp1[1] = t1 >> 1; temp1[2] = t1 >> 2;		\
+		puckU32ToThree_1(temp1[0]);		\
+		puckU32ToThree_1(temp1[1]);		\
+		puckU32ToThree_1(temp1[2]);		\
+		out[0] =  (temp1[0] << 10) | temp0[2];		\
+		out[1] =  (temp1[2] << 11) | temp0[1];		\
+		out[2] = (temp1[1] << 11) | temp0[0];		\
 }
 }
 #define U96_BIT_LOTR32_8(t0,t1,t2,t3,t4,t5){\
 #define U96_BIT_LOTR32_8(t0,t1,t2,t3,t4,t5){\
 t3= LOTR32(t2, 2);\
 t3= LOTR32(t2, 2);\

knot/Implementations/crypto_hash/knot384/armcortexm_3/hash.c

@@ -31,6 +31,10 @@ int crypto_hash(unsigned char *out, const unsigned char *in,
 	u32 s_temp[12] = { 0 };
 	u32 s_temp[12] = { 0 };
 	u8 i;
 	u8 i;
 	u32 dataFormat[3] = { 0 };
 	u32 dataFormat[3] = { 0 };
+	u32 t[3] = { 0 };
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
 	// initialization
 	// initialization
 	u32 s[12] = { 0 };
 	u32 s[12] = { 0 };
 	u8 tempData[12];
 	u8 tempData[12];

knot/Implementations/crypto_hash/knot384/armcortexm_4/auxFormat.c

 #include"auxFormat.h"
 #include"auxFormat.h"
 
 
-void unpackU96FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 }; \
-		u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-		t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) | \
-		((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) | \
-		((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-		t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10) | ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7) | \
-		((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5) | ((t2 & 0x7f800) >> 3) | \
-		((t0 & 0x3fc00) >> 10);		 \
-		t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1) | ((t2 & 0xc0000000) >> 2) | \
-		((t0 & 0x60000000) >> 3) | ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6) | \
-		((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);		 \
-		unpuckU32ToThree_3(t[0]);		 \
-		unpuckU32ToThree_3(t[1]);		 \
-		unpuckU32ToThree_3(t[2]);		 \
-		memcpy(out, t, 12 * sizeof(unsigned char));		 \
+void unpackU96FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t[3] = { 0 };
+	temp0[0] = in[2] & 0x7ff;
+	temp0[1] = in[1] & 0x7ff;
+	temp0[2] = in[0] & 0x3ff;
+	temp1[0] = (in[0]>>10) & 0x7ff;
+	temp1[1] = (in[2] >>11 ) & 0x7ff;
+	temp1[2] = (in[1] >> 11) & 0x3ff;
+	temp2[0] = in[1] >> 21;
+	temp2[1] = in[0] >> 21;
+	temp2[2] = in[2] >> 22;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;
+	unpuckU32ToThree_1(temp1[0]);
+	unpuckU32ToThree_1(temp1[1]);
+	unpuckU32ToThree_1(temp1[2]);
+	t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;
+	unpuckU32ToThree_1(temp2[0]);
+	unpuckU32ToThree_1(temp2[1]);
+	unpuckU32ToThree_1(temp2[2]);
+	t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;
+	memcpy(out, t, 12 * sizeof(unsigned char));
 }
 }
+
 void packU48FormatToThreePacket(u32 * out, u8 * in) {
 void packU48FormatToThreePacket(u32 * out, u8 * in) {
-	u32 t0 = U32BIG(*(u32*)(in)), t1 = (u32)U16BIG(*(u16*)(in + 4));	\
-		puckU32ToThree_3(t0);	\
-		puckU32ToThree_3(t1);	\
-		out[0] =  \
-		(((t1 & 0x40000000) >> 10) | ((t1 & 0x03000000) >> 6) | ((t1 & 0x000000ff) << 10)) | \
-		((((t0 & 0x30000000) >> 20) | ((t0 & 0xff0000) >> 16)));	\
-		out[1] =  \
-		((((t1 & 0x30000000) >> 9) | ((t1 & 0xff0000) >> 5)) | \
-		(((t0 & 0x80000000) >> 21) | ((t0 & 0xc000000) >> 18) | ((t0 & 0xff00) >> 8)));	\
-		out[2] =  \
-		(((t1 & 0x80000000) >> 10) | ((t1 & 0xc000000) >> 7) | ((t1 & 0xff00) << 3)) | \
-			(((t0 & 0x40000000) >> 20) | ((t0 & 0x03000000) >> 16) | (t0 & 0x000000ff));	\
+	u32 t1 = (u32)U16BIG(*(u16*)(in + 4));
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	temp0[0] = U32BIG(((u32*)in)[0]); temp0[1] = U32BIG(((u32*)in)[0]) >> 1; temp0[2] = U32BIG(((u32*)in)[0]) >> 2;
+	puckU32ToThree_1(temp0[0]);
+	puckU32ToThree_1(temp0[1]);
+	puckU32ToThree_1(temp0[2]);
+	temp1[0] = t1; temp1[1] = t1 >> 1; temp1[2] = t1 >> 2;
+	puckU32ToThree_1(temp1[0]);
+	puckU32ToThree_1(temp1[1]);
+	puckU32ToThree_1(temp1[2]);
+	out[0] =  (temp1[0] << 10) | temp0[2];
+	out[1] =  (temp1[2] << 11) | temp0[1];
+	out[2] = (temp1[1] << 11) | temp0[0];
 }
 }
 
 
 unsigned char  constant7Format[104] = {
 unsigned char  constant7Format[104] = {

knot/Implementations/crypto_hash/knot384/armcortexm_4/auxFormat.h

@@ -39,21 +39,17 @@ t4 = LOTR32(t2, 18);\
 t5 = LOTR32(t0, 19); \
 t5 = LOTR32(t0, 19); \
 }
 }
 unsigned char  constant7Format[104];
 unsigned char  constant7Format[104];
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-		u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }

knot/Implementations/crypto_hash/knot384/armcortexm_5/auxFormat.c

 #include"auxFormat.h"
 #include"auxFormat.h"
 
 
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-}
-#define unpuckU32ToThree_3(lo){\
-		u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-}
-void unpackU96FormatToThreePacket(u8 *out, u32 *in) {
-	u32 t[3] = { 0 }; \
-		u32 t0 = in[0], t1 = in[1], t2 = in[2];		 \
-		t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) | \
-		((t0 & 0x300) << 20) | ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) | \
-		((t0 & 0xff) << 16) | ((t1 & 0xff) << 8) | (t2 & 0xff);		 \
-		t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10) | ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7) | \
-		((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5) | ((t2 & 0x7f800) >> 3) | \
-		((t0 & 0x3fc00) >> 10);		 \
-		t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1) | ((t2 & 0xc0000000) >> 2) | \
-		((t0 & 0x60000000) >> 3) | ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6) | \
-		((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);		 \
-		unpuckU32ToThree_3(t[0]);		 \
-		unpuckU32ToThree_3(t[1]);		 \
-		unpuckU32ToThree_3(t[2]);		 \
-		memcpy(out, t, 12 * sizeof(unsigned char));		 \
+void unpackU96FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
+	u32 t[3] = { 0 };
+	temp0[0] = in[2] & 0x7ff;
+	temp0[1] = in[1] & 0x7ff;
+	temp0[2] = in[0] & 0x3ff;
+	temp1[0] = (in[0]>>10) & 0x7ff;
+	temp1[1] = (in[2] >>11 ) & 0x7ff;
+	temp1[2] = (in[1] >> 11) & 0x3ff;
+	temp2[0] = in[1] >> 21;
+	temp2[1] = in[0] >> 21;
+	temp2[2] = in[2] >> 22;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;
+	unpuckU32ToThree_1(temp1[0]);
+	unpuckU32ToThree_1(temp1[1]);
+	unpuckU32ToThree_1(temp1[2]);
+	t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;
+	unpuckU32ToThree_1(temp2[0]);
+	unpuckU32ToThree_1(temp2[1]);
+	unpuckU32ToThree_1(temp2[2]);
+	t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;
+	memcpy(out, t, 12 * sizeof(unsigned char));
 }
 }
+
 void packU48FormatToThreePacket(u32 * out, u8 * in) {
 void packU48FormatToThreePacket(u32 * out, u8 * in) {
-	u32 t0 = U32BIG(*(u32*)(in)), t1 = (u32)U16BIG(*(u16*)(in + 4));	\
-		puckU32ToThree_3(t0);	\
-		puckU32ToThree_3(t1);	\
-		out[0] =  \
-		(((t1 & 0x40000000) >> 10) | ((t1 & 0x03000000) >> 6) | ((t1 & 0x000000ff) << 10)) | \
-		((((t0 & 0x30000000) >> 20) | ((t0 & 0xff0000) >> 16)));	\
-		out[1] =  \
-		((((t1 & 0x30000000) >> 9) | ((t1 & 0xff0000) >> 5)) | \
-		(((t0 & 0x80000000) >> 21) | ((t0 & 0xc000000) >> 18) | ((t0 & 0xff00) >> 8)));	\
-		out[2] =  \
-		(((t1 & 0x80000000) >> 10) | ((t1 & 0xc000000) >> 7) | ((t1 & 0xff00) << 3)) | \
-			(((t0 & 0x40000000) >> 20) | ((t0 & 0x03000000) >> 16) | (t0 & 0x000000ff));	\
+	u32 t1 = (u32)U16BIG(*(u16*)(in + 4));
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	temp0[0] = U32BIG(((u32*)in)[0]); temp0[1] = U32BIG(((u32*)in)[0]) >> 1; temp0[2] = U32BIG(((u32*)in)[0]) >> 2;
+	puckU32ToThree_1(temp0[0]);
+	puckU32ToThree_1(temp0[1]);
+	puckU32ToThree_1(temp0[2]);
+	temp1[0] = t1; temp1[1] = t1 >> 1; temp1[2] = t1 >> 2;
+	puckU32ToThree_1(temp1[0]);
+	puckU32ToThree_1(temp1[1]);
+	puckU32ToThree_1(temp1[2]);
+	out[0] =  (temp1[0] << 10) | temp0[2];
+	out[1] =  (temp1[2] << 11) | temp0[1];
+	out[2] = (temp1[1] << 11) | temp0[0];
 }
 }
+
 unsigned char  constant7Format[104] = {
 unsigned char  constant7Format[104] = {
 	/*constant7Format[127]:  12*9=108*/
 	/*constant7Format[127]:  12*9=108*/
 0x01,0x08,0x40,0x02,0x10,0x80,0x05,0x09,0x48,0x42,0x12,0x90,
 0x01,0x08,0x40,0x02,0x10,0x80,0x05,0x09,0x48,0x42,0x12,0x90,

knot/Implementations/crypto_hash/knot384/armcortexm_6/auxFormat.c

 #include"auxFormat.h"
 #include"auxFormat.h"
 
 
-void unpackU96FormatToThreePacket(u8 *out, u32 *in) {
+void unpackU96FormatToThreePacket(u8 * out, u32 * in) {
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	u32 temp2[3] = { 0 };
 	u32 t[3] = { 0 };
 	u32 t[3] = { 0 };
-	u32 t0 = in[0], t1 = in[1], t2 = in[2];
-	t[0] = ((t1 & 0x400) << 21) | ((t2 & 0x400) << 20) | ((t0 & 0x300) << 20)
-			| ((t1 & 0x300) << 18) | ((t2 & 0x300) << 16) | ((t0 & 0xff) << 16)
-			| ((t1 & 0xff) << 8) | (t2 & 0xff);
-	t[1] = ((t2 & 0x200000) << 10) | ((t0 & 0x100000) << 10)
-			| ((t1 & 0x180000) << 9) | ((t2 & 0x180000) << 7)
-			| ((t0 & 0xc0000) << 6) | ((t1 & 0x7f800) << 5)
-			| ((t2 & 0x7f800) >> 3) | ((t0 & 0x3fc00) >> 10);
-	t[2] = ((t0 & 0x80000000)) | ((t1 & 0x80000000) >> 1)
-			| ((t2 & 0xc0000000) >> 2) | ((t0 & 0x60000000) >> 3)
-			| ((t1 & 0x60000000) >> 5) | ((t2 & 0x3fc00000) >> 6)
-			| ((t0 & 0x1fe00000) >> 13) | ((t1 & 0x1fe00000) >> 21);
-	unpuckU32ToThree_3(t[0]);
-	unpuckU32ToThree_3(t[1]);
-	unpuckU32ToThree_3(t[2]);
+	temp0[0] = in[2] & 0x7ff;
+	temp0[1] = in[1] & 0x7ff;
+	temp0[2] = in[0] & 0x3ff;
+	temp1[0] = (in[0]>>10) & 0x7ff;
+	temp1[1] = (in[2] >>11 ) & 0x7ff;
+	temp1[2] = (in[1] >> 11) & 0x3ff;
+	temp2[0] = in[1] >> 21;
+	temp2[1] = in[0] >> 21;
+	temp2[2] = in[2] >> 22;
+	unpuckU32ToThree_1(temp0[0]);
+	unpuckU32ToThree_1(temp0[1]);
+	unpuckU32ToThree_1(temp0[2]);
+	t[0] = temp0[0] | temp0[1] << 1 | temp0[2] << 2;
+	unpuckU32ToThree_1(temp1[0]);
+	unpuckU32ToThree_1(temp1[1]);
+	unpuckU32ToThree_1(temp1[2]);
+	t[1] = temp1[0] | temp1[1] << 1 | temp1[2] << 2;
+	unpuckU32ToThree_1(temp2[0]);
+	unpuckU32ToThree_1(temp2[1]);
+	unpuckU32ToThree_1(temp2[2]);
+	t[2] = temp2[0] | temp2[1] << 1 | temp2[2] << 2;
 	memcpy(out, t, 12 * sizeof(unsigned char));
 	memcpy(out, t, 12 * sizeof(unsigned char));
 }
 }
-void packU48FormatToThreePacket(u32 *out, u8 *in) {
-	u32 t0 = U32BIG(*(u32* )(in)), t1 = (u32) U16BIG(*(u16* )(in + 4));
-	puckU32ToThree_3(t0);
-	puckU32ToThree_3(t1);
-	out[0] = (((t1 & 0x40000000) >> 10) | ((t1 & 0x03000000) >> 6)
-			| ((t1 & 0x000000ff) << 10))
-			| ((((t0 & 0x30000000) >> 20) | ((t0 & 0xff0000) >> 16)));
-	out[1] = ((((t1 & 0x30000000) >> 9) | ((t1 & 0xff0000) >> 5))
-			| (((t0 & 0x80000000) >> 21) | ((t0 & 0xc000000) >> 18)
-					| ((t0 & 0xff00) >> 8)));
-	out[2] = (((t1 & 0x80000000) >> 10) | ((t1 & 0xc000000) >> 7)
-			| ((t1 & 0xff00) << 3))
-			| (((t0 & 0x40000000) >> 20) | ((t0 & 0x03000000) >> 16)
-					| (t0 & 0x000000ff));
-}
 
 
+void packU48FormatToThreePacket(u32 * out, u8 * in) {
+	u32 t1 = (u32)U16BIG(*(u16*)(in + 4));
+	u32 temp0[3] = { 0 };
+	u32 temp1[3] = { 0 };
+	temp0[0] = U32BIG(((u32*)in)[0]); temp0[1] = U32BIG(((u32*)in)[0]) >> 1; temp0[2] = U32BIG(((u32*)in)[0]) >> 2;
+	puckU32ToThree_1(temp0[0]);
+	puckU32ToThree_1(temp0[1]);
+	puckU32ToThree_1(temp0[2]);
+	temp1[0] = t1; temp1[1] = t1 >> 1; temp1[2] = t1 >> 2;
+	puckU32ToThree_1(temp1[0]);
+	puckU32ToThree_1(temp1[1]);
+	puckU32ToThree_1(temp1[2]);
+	out[0] =  (temp1[0] << 10) | temp0[2];
+	out[1] =  (temp1[2] << 11) | temp0[1];
+	out[2] = (temp1[1] << 11) | temp0[0];
+}
 unsigned char constant7Format[104] = {
 unsigned char constant7Format[104] = {
 /*constant7Format[127]:  12*9=108*/
 /*constant7Format[127]:  12*9=108*/
 0x01, 0x08, 0x40, 0x02, 0x10, 0x80, 0x05, 0x09, 0x48, 0x42, 0x12, 0x90, 0x85,
 0x01, 0x08, 0x40, 0x02, 0x10, 0x80, 0x05, 0x09, 0x48, 0x42, 0x12, 0x90, 0x85,

knot/Implementations/crypto_hash/knot384/armcortexm_6/auxFormat.h

@@ -12,24 +12,19 @@ typedef unsigned short u16;
 typedef unsigned int u32;
 typedef unsigned int u32;
 typedef unsigned long long u64;
 typedef unsigned long long u64;
 
 
-
-#define puckU32ToThree_3(lo){\
-u32 r0;\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
+#define puckU32ToThree_1(x){\
+x &= 0x49249249;\
+x = (x | (x >>  2)) & 0xc30c30c3;\
+x = (x | (x >>4)) & 0x0f00f00f;\
+x = (x | (x >> 8)) & 0xff0000ff;\
+x = (x | (x >> 16)) & 0xfff;\
 }
 }
-#define unpuckU32ToThree_3(lo){\
-		u32 r0;\
-r0 = (lo ^ (lo << 12)) & 0x000f0000, lo ^= r0 ^ (r0 >> 12);\
-r0 = (lo ^ (lo << 4)) & 0x000f0f00, lo ^= r0 ^ (r0 >> 4);\
-r0 = (lo ^ (lo << 6)) & 0x00300300, lo ^= r0 ^ (r0 >> 6);\
-r0 = (lo ^ (lo << 2)) & 0x00330330, lo ^= r0 ^ (r0 >> 2);\
-r0 = (lo ^ (lo << 3)) & 0x10410410, lo ^= r0 ^ (r0 >> 3);\
-r0 = (lo ^ (lo << 1)) & 0x14514514, lo ^= r0 ^ (r0 >> 1);\
+#define unpuckU32ToThree_1(x){\
+x &= 0xfff;\
+x = (x | (x << 16)) & 0xff0000ff;\
+x = (x | (x << 8)) & 0x0f00f00f;\
+x = (x | (x << 4)) & 0xc30c30c3;\
+x = (x | (x << 2)) & 0x49249249;\
 }
 }
 #define ARC(rci) \
 #define ARC(rci) \
   do { \
   do { \

knot/Implementations/crypto_hash/knot384/armcortexm_6/hash.c

@@ -8,7 +8,7 @@
 
 
 int crypto_hash(unsigned char *out, const unsigned char *in,
 int crypto_hash(unsigned char *out, const unsigned char *in,
 		unsigned long long inlen) {
 		unsigned long long inlen) {
-	u32 dataFormat[3] = { 0 }, t1, t2;
+	u32 dataFormat[3] = { 0 }, t2;
 	// initialization
 	// initialization
 	u32 s[12] = { 0 };
 	u32 s[12] = { 0 };
 	u8 tempData[12];
 	u8 tempData[12];