diff --git a/knot/Implementations/crypto_aead/knot128v1/armcortexm_1/encrypt.c b/knot/Implementations/crypto_aead/knot128v1/armcortexm_1/encrypt.c
index a7fd7a4..2cb824b 100644
--- a/knot/Implementations/crypto_aead/knot128v1/armcortexm_1/encrypt.c
+++ b/knot/Implementations/crypto_aead/knot128v1/armcortexm_1/encrypt.c
@@ -59,87 +59,93 @@ static void permutation256(unsigned char *in, int rounds, unsigned char *rc) {
 	uint32_t one = 0x1;
 	uint32_t ff = 0xff;
 	__asm volatile(
-		"ldr     w0,     [in]          \n\t"
-		"ldr     w4,     [in, #4]      \n\t"
-		"ldr     w1,     [in, #8]      \n\t"
-		"ldr     w5,     [in, #12]     \n\t"
-		"ldr     w2,     [in, #16]     \n\t"
-		"ldr     w6,     [in, #20]     \n\t"
-		"ldr     w3,     [in, #24]     \n\t"
-		"ldr     w7,     [in, #28]     \n\t"
-		"mov     s0,     0xfff         \n\t"
-		"mov     s2,     0x1fff        \n\t"
-		"lsl     s2,     s2, #12       \n\t"
-		"eors    s2,     s2, s0        \n\t"
-	"enc_loop:                       \n\t"
+	"enc_loop_%=:                           \n\t"
+		"ldr     %[w0],     [%[in]]         \n\t"
+		"ldr     %[w4],     [%[in], #4]     \n\t"
+		"ldr     %[w1],     [%[in], #8]     \n\t"
+		"ldr     %[w5],     [%[in], #12]    \n\t"
+		"ldr     %[w2],     [%[in], #16]    \n\t"
+		"ldr     %[w6],     [%[in], #20]    \n\t"
+		"ldr     %[w3],     [%[in], #24]    \n\t"
+		"ldr     %[w7],     [%[in], #28]    \n\t"
+		"mov     %[s0],     0xfff           \n\t"
+		"mov     %[s2],     0x1fff          \n\t"
+		"lsl     %[s2],     %[s2], #12      \n\t"
+		"eors    %[s2],     %[s2], %[s0]    \n\t"
     "/*add round const*/           \n\t"
-		"ldrb    s0,     [rc]          \n\t"
-	  "eors    w0,     w0, s0        \n\t"
+		"ldrb    %[s0],     [%[rc]]         \n\t"
+	    "eors    %[w0],     %[w0], %[s0]    \n\t"
     "/*sbox first column*/         \n\t"
-		"mvns    w0,     w0            \n\t"
-		"ands    s0,     w1, w0        \n\t"
-		"eors    s0,     w2, s0        \n\t"
-		"orrs    w2,     w1, w2        \n\t"
-		"eors    w0,     w3, w0        \n\t"
-		"eors    w2,     w2, w0        \n\t"
-		"eors    s1,     w1, w3        \n\t"
-		"eors    w3,     w3, s0        \n\t"
-		"ands    w0,     s0, w0        \n\t"
-		"eors    w0,     s1, w0        \n\t"
-		"ands    w1,     w2, s1        \n\t"
-		"eors    w1,     s0, w1        \n\t"
+		"mvns    %[w0],     %[w0]           \n\t"
+		"ands    %[s0],     %[w1], %[w0]    \n\t"
+		"eors    %[s0],     %[w2], %[s0]    \n\t"
+		"orrs    %[w2],     %[w1], %[w2]    \n\t"
+		"eors    %[w0],     %[w3], %[w0]    \n\t"
+		"eors    %[w2],     %[w2], %[w0]    \n\t"
+		"eors    %[s1],     %[w1], %[w3]    \n\t"
+		"eors    %[w3],     %[w3], %[s0]    \n\t"
+		"ands    %[w0],     %[s0], %[w0]    \n\t"
+		"eors    %[w0],     %[s1], %[w0]    \n\t"
+		"ands    %[w1],     %[w2], %[s1]    \n\t"
+		"eors    %[w1],     %[s0], %[w1]    \n\t"
 		"/*sbox second column*/        \n\t"
-		"mvns    w4,     w4            \n\t"
-		"ands    s0,     w5, w4        \n\t"
-		"eors    s0,     w6, s0        \n\t"
-		"orrs    w6,     w5, w6        \n\t"
-		"eors    w4,     w7, w4        \n\t"
-		"eors    w6,     w6, w4        \n\t"
-		"eors    s1,     w5, w7        \n\t"
-		"eors    w7,     w7, s0        \n\t"
-		"ands    w4,     s0, w4        \n\t"
-		"eors    w4,     s1, w4        \n\t"
-		"ands    w5,     w6, s1        \n\t"
-		"eors    w5,     s0, w5        \n\t"
+		"mvns    %[w4],     %[w4]           \n\t"
+		"ands    %[s0],     %[w5], %[w4]    \n\t"
+		"eors    %[s0],     %[w6], %[s0]    \n\t"
+		"orrs    %[w6],     %[w5], %[w6]    \n\t"
+		"eors    %[w4],     %[w7], %[w4]    \n\t"
+		"eors    %[w6],     %[w6], %[w4]    \n\t"
+		"eors    %[s1],     %[w5], %[w7]    \n\t"
+		"eors    %[w7],     %[w7], %[s0]    \n\t"
+		"ands    %[w4],     %[s0], %[w4]    \n\t"
+		"eors    %[w4],     %[s1], %[w4]    \n\t"
+		"ands    %[w5],     %[w6], %[s1]    \n\t"
+		"eors    %[w5],     %[s0], %[w5]    \n\t"
     "/*rotate shift left 1 bit*/   \n\t"
-		"ror     s0,     w1, #31       \n\t"
-		"ands    s0,     s0, one       \n\t"
-		"lsl     w1,     w1, #1        \n\t"
-		"ror     s1,     w5, #31       \n\t"
-		"ands    s1,     s1, one       \n\t"
-		"eors    w1,     w1, s1        \n\t"
-		"lsl     w5,     w5, #1        \n\t"
-		"eors    w5,     w5, s0        \n\t"
+		"ror     %[s0],     %[w1], #31      \n\t"
+		"ands    %[s0],     %[s0], %[one]   \n\t"
+		"lsl     %[w1],     %[w1], #1       \n\t"
+		"ror     %[s1],     %[w5], #31      \n\t"
+		"ands    %[s1],     %[s1], %[one]   \n\t"
+		"eors    %[w1],     %[w1], %[s1]    \n\t"
+		"lsl     %[w5],     %[w5], #1       \n\t"
+		"eors    %[w5],     %[w5], %[s0]    \n\t"
     "/*rotate shift left 8 bits*/  \n\t"
-		"ror     s0,     w2, #24       \n\t"
-		"ands    s0,     s0, ff        \n\t"
-		"lsl     w2,     w2, #8        \n\t"
-		"ror     s1,     w6, #24       \n\t"
-		"ands    s1,     s1, ff        \n\t"
-		"eors    w2,     w2, s1        \n\t"
-		"lsl     w6,     w6, #8        \n\t"
-		"eors    w6,     w6, s0        \n\t"
+		"ror     %[s0],     %[w2], #24      \n\t"
+		"ands    %[s0],     %[s0], %[ff]    \n\t"
+		"lsl     %[w2],     %[w2], #8       \n\t"
+		"ror     %[s1],     %[w6], #24      \n\t"
+		"ands    %[s1],     %[s1], %[ff]    \n\t"
+		"eors    %[w2],     %[w2], %[s1]    \n\t"
+		"lsl     %[w6],     %[w6], #8       \n\t"
+		"eors    %[w6],     %[w6], %[s0]    \n\t"
     "/*rotate shift left 25 bits*/ \n\t"
-		"ror     s0,     w3, #7        \n\t"
-		"ands    s0,     s0, s2        \n\t"
-		"lsl     w3,     w3, #25       \n\t"
-		"ror     s1,     w7, #7        \n\t"
-		"ands    s1,     s1, s2        \n\t"
-		"eors    w3,     w3, s1        \n\t"
-		"lsl     w7,     w7, #25       \n\t"
-		"eors    w7,     w7, s0        \n\t"
+		"ror     %[s0],     %[w3], #7       \n\t"
+		"ands    %[s0],     %[s0], %[s2]    \n\t"
+		"lsl     %[w3],     %[w3], #25      \n\t"
+		"ror     %[s1],     %[w7], #7       \n\t"
+		"ands    %[s1],     %[s1], %[s2]    \n\t"
+		"eors    %[w3],     %[w3], %[s1]    \n\t"
+		"lsl     %[w7],     %[w7], #25      \n\t"
+		"eors    %[w7],     %[w7], %[s0]    \n\t"
 		"/*loop control*/              \n\t"
- 		"adds    rc,     rc, #1        \n\t"
-		"subs    rounds, rounds,  #1   \n\t"
-		"bne     enc_loop              \n\t"
-		"str     w0,     [in]         \n\t"
-		"str     w4,     [in, #4]     \n\t"
-		"str     w1,     [in, #8]     \n\t"
-		"str     w5,     [in, #12]    \n\t"
-		"str     w2,     [in, #16]    \n\t"
-		"str     w6,     [in, #20]    \n\t"
-		"str     w3,     [in, #24]    \n\t"
-		"str     w7,     [in, #28]    \n\t"
+		"adds    %[rc],     %[rc], #1       \n\t"
+		"subs    %[rounds], %[rounds],  #1  \n\t"
+		"bne     enc_loop_%=                \n\t"
+		"str     %[w0],     [%[in]]         \n\t"
+		"str     %[w4],     [%[in], #4]     \n\t"
+		"str     %[w1],     [%[in], #8]     \n\t"
+		"str     %[w5],     [%[in], #12]    \n\t"
+		"str     %[w2],     [%[in], #16]    \n\t"
+		"str     %[w6],     [%[in], #20]    \n\t"
+		"str     %[w3],     [%[in], #24]    \n\t"
+		"str     %[w7],     [%[in], #28]    \n\t"
+
+        : [rounds] "=r" (rounds), [rc] "=r" (rc),
+          [w0] "=r" (w0), [w1] "=r" (w1), [w2] "=r" (w2), [w3] "=r" (w3),
+          [w4] "=r" (w4), [w5] "=r" (w5), [w6] "=r" (w6), [w7] "=r" (w7),
+          [s0] "=r" (s0), [s1] "=r" (s1), [s2] "=r" (s2)
+        : [in] "r" (in), "[rounds]" (rounds), "[rc]" (rc), [ff] "r" (ff), [one] "r" (one)
 	);
 }
 
diff --git a/knot/Implementations/crypto_aead/knot128v1/armcortexm_2/encrypt.c b/knot/Implementations/crypto_aead/knot128v1/armcortexm_2/encrypt.c
index 4b84924..56d4962 100644
--- a/knot/Implementations/crypto_aead/knot128v1/armcortexm_2/encrypt.c
+++ b/knot/Implementations/crypto_aead/knot128v1/armcortexm_2/encrypt.c
@@ -80,74 +80,77 @@ unsigned char  constant6Format[63] = {
 static void permutation256(unsigned int *in, int rounds, unsigned char *rc) {
 	uint32_t w0, w1, w2, w3, w4, w5, w6, w7;
 	uint32_t s0, s1, s2;
-	uint32_t one = 0x1;
-	uint32_t i=0;
-	uint32_t ff = 0xff;
 	__asm volatile(
-		"ldr     w0,     [in]          \n\t"
-		"ldr     w4,     [in, #4]      \n\t"
-		"ldr     w1,     [in, #8]      \n\t"
-		"ldr     w5,     [in, #12]     \n\t"
-		"ldr     w2,     [in, #16]     \n\t"
-		"ldr     w6,     [in, #20]     \n\t"
-		"ldr     w3,     [in, #24]     \n\t"
-		"ldr     w7,     [in, #28]     \n\t"	
-	"enc_loop:                       \n\t"
+		"ldr     %[w0],     [%[in]]          \n\t"
+		"ldr     %[w4],     [%[in], #4]      \n\t"
+		"ldr     %[w1],     [%[in], #8]      \n\t"
+		"ldr     %[w5],     [%[in], #12]     \n\t"
+		"ldr     %[w2],     [%[in], #16]     \n\t"
+		"ldr     %[w6],     [%[in], #20]     \n\t"
+		"ldr     %[w3],     [%[in], #24]     \n\t"
+		"ldr     %[w7],     [%[in], #28]     \n\t"
+	"enc_loop_%=:                       \n\t"
     "/*add round const   s0 s1*/           \n\t"
-		"ldrb    s0,     [rc]          \n\t"	
-		"LSR     s1,     s0, #4       \n\t"	
-		"and    s0,     s0, 0xf        \n\t"
-	  "eors    w4,     w4, s0        \n\t"
-	  "eors    w0,     w0, s1        \n\t"
+		"ldrb    %[s0],     [%[rc]]          \n\t"
+		"LSR     %[s1],     %[s0], #4       \n\t"
+		"and    %[s0],     %[s0], 0xf        \n\t"
+	  "eors    %[w4],     %[w4], %[s0]        \n\t"
+	  "eors    %[w0],     %[w0], %[s1]        \n\t"
     "/*sbox first column*/         \n\t"
-		"mvns    w0,     w0            \n\t"
-		"ands    s0,     w1, w0        \n\t"
-		"eors    s0,     w2, s0        \n\t"
-		"orrs    w2,     w1, w2        \n\t"
-		"eors    w0,     w3, w0        \n\t"
-		"eors    w2,     w2, w0        \n\t"
-		"eors    s1,     w1, w3        \n\t"
-		"eors    w3,     w3, s0        \n\t"
-		"ands    w0,     s0, w0        \n\t"
-		"eors    w0,     s1, w0        \n\t"
-		"ands    w1,     w2, s1        \n\t"
-		"eors    w1,     s0, w1        \n\t"
+		"mvns    %[w0],     %[w0]            \n\t"
+		"ands    %[s0],     %[w1], %[w0]        \n\t"
+		"eors    %[s0],     %[w2], %[s0]        \n\t"
+		"orrs    %[w2],     %[w1], %[w2]        \n\t"
+		"eors    %[w0],     %[w3], %[w0]        \n\t"
+		"eors    %[w2],     %[w2], %[w0]        \n\t"
+		"eors    %[s1],     %[w1], %[w3]        \n\t"
+		"eors    %[w3],     %[w3], %[s0]        \n\t"
+		"ands    %[w0],     %[s0], %[w0]        \n\t"
+		"eors    %[w0],     %[s1], %[w0]        \n\t"
+		"ands    %[w1],     %[w2], %[s1]        \n\t"
+		"eors    %[w1],     %[s0], %[w1]        \n\t"
 		"/*sbox second column*/        \n\t"
-		"mvns    w4,     w4            \n\t"
-		"ands    s0,     w5, w4        \n\t"
-		"eors    s0,     w6, s0        \n\t"
-		"orrs    w6,     w5, w6        \n\t"
-		"eors    w4,     w7, w4        \n\t"
-		"eors    w6,     w6, w4        \n\t"
-		"eors    s1,     w5, w7        \n\t"
-		"eors    w7,     w7, s0        \n\t"
-		"ands    w4,     s0, w4        \n\t"
-		"eors    w4,     s1, w4        \n\t"
-		"ands    w5,     w6, s1        \n\t"
-		"eors    w5,     s0, w5        \n\t"
+		"mvns    %[w4],     %[w4]            \n\t"
+		"ands    %[s0],     %[w5], %[w4]        \n\t"
+		"eors    %[s0],     %[w6], %[s0]        \n\t"
+		"orrs    %[w6],     %[w5], %[w6]        \n\t"
+		"eors    %[w4],     %[w7], %[w4]        \n\t"
+		"eors    %[w6],     %[w6], %[w4]        \n\t"
+		"eors    %[s1],     %[w5], %[w7]        \n\t"
+		"eors    %[w7],     %[w7], %[s0]        \n\t"
+		"ands    %[w4],     %[s0], %[w4]        \n\t"
+		"eors    %[w4],     %[s1], %[w4]        \n\t"
+		"ands    %[w5],     %[w6], %[s1]        \n\t"
+		"eors    %[w5],     %[s0], %[w5]        \n\t"
     "/*rotate shift left 1 bit*/   \n\t"
-		"mov    s0,     w5       \n\t"
-		"ROR    w5,     w1, #31        \n\t"
-		"mov    w1,     s0       \n\t"
+		"mov    %[s0],     %[w5]       \n\t"
+		"ROR    %[w5],     %[w1], #31        \n\t"
+		"mov    %[w1],     %[s0]       \n\t"
     "/*rotate shift left 8 bits*/  \n\t"
-		"ROR    w2,     w2, #28        \n\t"
-		"ROR    w6,     w6, #28       \n\t"
+		"ROR    %[w2],     %[w2], #28        \n\t"
+		"ROR    %[w6],     %[w6], #28       \n\t"
     "/*rotate shift left 25 bits*/ \n\t"
-		"mov    s0,     w3       \n\t"
-		"ROR    w3,     w7, #20       \n\t"
-		"ROR    w7,     s0, #19       \n\t"
+		"mov    %[s0],     %[w3]       \n\t"
+		"ROR    %[w3],     %[w7], #20       \n\t"
+		"ROR    %[w7],     %[s0], #19       \n\t"
 		"/*loop control*/              \n\t"
- 		"adds    rc,     rc, #1        \n\t"
-		"subs    rounds, rounds,  #1   \n\t"
-		"bne     enc_loop              \n\t"
-		"str     w0,     [in]         \n\t"
-		"str     w4,     [in, #4]     \n\t"
-		"str     w1,     [in, #8]     \n\t"
-		"str     w5,     [in, #12]    \n\t"
-		"str     w2,     [in, #16]    \n\t"
-		"str     w6,     [in, #20]    \n\t"
-		"str     w3,     [in, #24]    \n\t"
-		"str     w7,     [in, #28]    \n\t"
+		"adds    %[rc],     %[rc], #1        \n\t"
+		"subs    %[rounds], %[rounds],  #1   \n\t"
+		"bne     enc_loop_%=              \n\t"
+		"str     %[w0],     [%[in]]         \n\t"
+		"str     %[w4],     [%[in], #4]     \n\t"
+		"str     %[w1],     [%[in], #8]     \n\t"
+		"str     %[w5],     [%[in], #12]    \n\t"
+		"str     %[w2],     [%[in], #16]    \n\t"
+		"str     %[w6],     [%[in], #20]    \n\t"
+		"str     %[w3],     [%[in], #24]    \n\t"
+		"str     %[w7],     [%[in], #28]    \n\t"
+
+        : [rounds] "=r" (rounds), [rc] "=r" (rc),
+          [w0] "=r" (w0), [w1] "=r" (w1), [w2] "=r" (w2), [w3] "=r" (w3),
+          [w4] "=r" (w4), [w5] "=r" (w5), [w6] "=r" (w6), [w7] "=r" (w7),
+          [s0] "=r" (s0), [s1] "=r" (s1), [s2] "=r" (s2)
+        : [in] "r" (in), "[rounds]" (rounds), "[rc]" (rc)
 	);
 }