/*************************************************************************/
// #define CHANGE_ENDIAN32(x) (((x)<<24) | ((x)>>24) | (((x)& 0x0000ff00)<<8) | (((x)& 0x00ff0000)>>8))
+/*
static
uint32_t change_endian32(uint32_t x){
return (((x)<<24) | ((x)>>24) | (((x)& 0x0000ff00)<<8) | (((x)& 0x00ff0000)>>8));
}
-
+*/
/* sha256 functions as macros for speed and size, cause they are called only once */
#define CH(x,y,z) (((x)&(y)) ^ ((~(x))&(z)))
#define MAJ(x,y,z) (((x)&(y)) ^ ((x)&(z)) ^ ((y)&(z)))
-#define SIGMA0(x) (rotr32((x),2) ^ rotr32((x),13) ^ rotl32((x),10))
-#define SIGMA1(x) (rotr32((x),6) ^ rotr32((x),11) ^ rotl32((x),7))
-#define SIGMA_a(x) (rotr32((x),7) ^ rotl32((x),14) ^ ((x)>>3))
+#define SIGMA_0(x) (rotr32((x), 2) ^ rotr32((x),13) ^ rotl32((x),10))
+#define SIGMA_1(x) (rotr32((x), 6) ^ rotr32((x),11) ^ rotl32((x),7))
+#define SIGMA_a(x) (rotr32((x), 7) ^ rotl32((x),14) ^ ((x)>>3))
#define SIGMA_b(x) (rotl32((x),15) ^ rotl32((x),13) ^ ((x)>>10))
const
0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208, 0x90befffa, 0xa4506ceb, 0xbef9a3f7, 0xc67178f2
};
-
+static
+void load_endian32_changed(uint8_t* dest, uint8_t* src, uint16_t words){
+ while(words--){
+ *dest++ = src[3];
+ *dest++ = src[2];
+ *dest++ = src[1];
+ *dest++ = src[0];
+ src += 4;
+ }
+}
/**
* block must be, 512, Bit = 64, Byte, long !!!
*/
void sha2_small_common_nextBlock (sha2_small_common_ctx_t *state, const void* block){
- uint32_t w[64]; /* this is 256, byte, large, */
+ uint32_t w[16], wx;
uint8_t i;
uint32_t a[8],t1,t2;
/* init w */
#if defined LITTLE_ENDIAN
- for (i=0; i<16; ++i){
- w[i]= change_endian32(((uint32_t*)block)[i]);
- }
+ load_endian32_changed((uint8_t*)w, (uint8_t*)block, 16);
#elif defined BIG_ENDIAN
memcpy((void*)w, block, 64);
#endif
- for (i=16; i<64; ++i){
- w[i] = SIGMA_b(w[i-2]) + w[i-7] + SIGMA_a(w[i-15]) + w[i-16];
- }
-
- /* init working variables */
- memcpy((void*)a,(void*)(state->h), 8*4);
-
- /* do the, fun stuff, */
- for (i=0; i<64; ++i){
- t1 = a[7] + SIGMA1(a[4]) + CH(a[4],a[5],a[6]) + k[i] + w[i];
- t2 = SIGMA0(a[0]) + MAJ(a[0],a[1],a[2]);
- memmove(&(a[1]), &(a[0]), 7*4); /* a[7]=a[6]; a[6]=a[5]; a[5]=a[4]; a[4]=a[3]; a[3]=a[2]; a[2]=a[1]; a[1]=a[0]; */
- a[4] += t1;
- a[0] = t1 + t2;
+/*
+ for (i=16; i<64; ++i){
+ w[i] = SIGMA_b(w[i-2]) + w[i-7] + SIGMA_a(w[i-15]) + w[i-16];
+ }
+*/
+/* init working variables */
+ memcpy((void*)a,(void*)(state->h), 8*4);
+
+/* do the, fun stuff, */
+ for (i=0; i<64; ++i){
+ if(i<16){
+ wx = w[i];
+ }else{
+ wx = SIGMA_b(w[14]) + w[9] + SIGMA_a(w[1]) + w[0];
+ memmove(&(w[0]), &(w[1]), 15*4);
+ w[15] = wx;
}
+ t1 = a[7] + SIGMA_1(a[4]) + CH(a[4],a[5],a[6]) + k[i] + wx;
+ t2 = SIGMA_0(a[0]) + MAJ(a[0],a[1],a[2]);
+ memmove(&(a[1]), &(a[0]), 7*4); /* a[7]=a[6]; a[6]=a[5]; a[5]=a[4]; a[4]=a[3]; a[3]=a[2]; a[2]=a[1]; a[1]=a[0]; */
+ a[4] += t1;
+ a[0] = t1 + t2;
+ }
- /* update, the, state, */
- for (i=0; i<8; ++i){
- state->h[i] += a[i];
- }
- state->length += 1;
+/* update, the, state, */
+ for (i=0; i<8; ++i){
+ state->h[i] += a[i];
+ }
+ state->length += 1;
}
/* set the final one bit */
lb[length_b/8] |= 0x80>>(length_b & 0x7);
/* pad with zeros */
- if (length_b>512-64){ /* not enouth space for 64bit length value */
+ if (length_b>=512-64){ /* not enouth space for 64bit length value */
sha2_small_common_nextBlock(state, lb);
memset(lb, 0, 64);
}