Default transform

void SHA2_SHA256_Transform(SHA2_SHA256_CTX* context, const sha2_word32* data) {
	sha2_word32	a, b, c, d, e, f, g, h, s0, s1;
	sha2_word32	T1, T2, *W256;
	int		j;

	W256 = (sha2_word32*)context->buffer;

	/* Initialize registers with the prev. intermediate value */
	a = context->state[0];
	b = context->state[1];
	c = context->state[2];
	d = context->state[3];
	e = context->state[4];
	f = context->state[5];
	g = context->state[6];
	h = context->state[7];

	j = 0;
	do {
#if BYTE_ORDER == LITTLE_ENDIAN
		/* Copy data while converting to host byte order */
		REVERSE32(*data++,W256[j]);
		/* Apply the SHA-256 compression function to update a..h */
		T1 = h + Sigma1_256(e) + Ch(e, f, g) + K256[j] + W256[j];
#else /* BYTE_ORDER == LITTLE_ENDIAN */
		/* Apply the SHA-256 compression function to update a..h with copy */
		T1 = h + Sigma1_256(e) + Ch(e, f, g) + K256[j] + (W256[j] = *data++);
#endif /* BYTE_ORDER == LITTLE_ENDIAN */
		T2 = Sigma0_256(a) + Maj(a, b, c);
		h = g;
		g = f;
		f = e;
		e = d + T1;
		d = c;
		c = b;
		b = a;
		a = T1 + T2;

		j++;
	} while (j < 16);

	do {
		/* Part of the message block expansion: */
		s0 = W256[(j+1)&0x0f];
		s0 = sigma0_256(s0);
		s1 = W256[(j+14)&0x0f];	
		s1 = sigma1_256(s1);

		/* Apply the SHA-256 compression function to update a..h */
		T1 = h + Sigma1_256(e) + Ch(e, f, g) + K256[j] + 
		     (W256[j&0x0f] += s1 + W256[(j+9)&0x0f] + s0);
		T2 = Sigma0_256(a) + Maj(a, b, c);
		h = g;
		g = f;
		f = e;
		e = d + T1;
		d = c;
		c = b;
		b = a;
		a = T1 + T2;

		j++;
	} while (j < 64);

	/* Compute the current intermediate hash value */
	context->state[0] += a;
	context->state[1] += b;
	context->state[2] += c;
	context->state[3] += d;
	context->state[4] += e;
	context->state[5] += f;
	context->state[6] += g;
	context->state[7] += h;

	/* Clean up */
	a = b = c = d = e = f = g = h = T1 = T2 = 0;
}

Unrolled transform

#ifdef SHA2_UNROLL_TRANSFORM

/* Unrolled SHA-256 round macros: */

#if BYTE_ORDER == LITTLE_ENDIAN

#define ROUND256_0_TO_15(a,b,c,d,e,f,g,h)	\
	REVERSE32(*data++, W256[j]); \
	T1 = (h) + Sigma1_256(e) + Ch((e), (f), (g)) + \
             K256[j] + W256[j]; \
	(d) += T1; \
	(h) = T1 + Sigma0_256(a) + Maj((a), (b), (c)); \
	j++


#else /* BYTE_ORDER == LITTLE_ENDIAN */

#define ROUND256_0_TO_15(a,b,c,d,e,f,g,h)	\
	T1 = (h) + Sigma1_256(e) + Ch((e), (f), (g)) + \
	     K256[j] + (W256[j] = *data++); \
	(d) += T1; \
	(h) = T1 + Sigma0_256(a) + Maj((a), (b), (c)); \
	j++

#endif /* BYTE_ORDER == LITTLE_ENDIAN */

#define ROUND256(a,b,c,d,e,f,g,h)	\
	s0 = W256[(j+1)&0x0f]; \
	s0 = sigma0_256(s0); \
	s1 = W256[(j+14)&0x0f]; \
	s1 = sigma1_256(s1); \
	T1 = (h) + Sigma1_256(e) + Ch((e), (f), (g)) + K256[j] + \
	     (W256[j&0x0f] += s1 + W256[(j+9)&0x0f] + s0); \
	(d) += T1; \
	(h) = T1 + Sigma0_256(a) + Maj((a), (b), (c)); \
	j++

void SHA2_SHA256_Transform(SHA2_SHA256_CTX* context, const sha2_word32* data) {
	sha2_word32	a, b, c, d, e, f, g, h, s0, s1;
	sha2_word32	T1, *W256;
	int		j;

	W256 = (sha2_word32*)context->buffer;

	/* Initialize registers with the prev. intermediate value */
	a = context->state[0];
	b = context->state[1];
	c = context->state[2];
	d = context->state[3];
	e = context->state[4];
	f = context->state[5];
	g = context->state[6];
	h = context->state[7];

	j = 0;
	do {
		/* Rounds 0 to 15 (unrolled): */
		ROUND256_0_TO_15(a,b,c,d,e,f,g,h);
		ROUND256_0_TO_15(h,a,b,c,d,e,f,g);
		ROUND256_0_TO_15(g,h,a,b,c,d,e,f);
		ROUND256_0_TO_15(f,g,h,a,b,c,d,e);
		ROUND256_0_TO_15(e,f,g,h,a,b,c,d);
		ROUND256_0_TO_15(d,e,f,g,h,a,b,c);
		ROUND256_0_TO_15(c,d,e,f,g,h,a,b);
		ROUND256_0_TO_15(b,c,d,e,f,g,h,a);
	} while (j < 16);

	/* Now for the remaining rounds to 64: */
	do {
		ROUND256(a,b,c,d,e,f,g,h);
		ROUND256(h,a,b,c,d,e,f,g);
		ROUND256(g,h,a,b,c,d,e,f);
		ROUND256(f,g,h,a,b,c,d,e);
		ROUND256(e,f,g,h,a,b,c,d);
		ROUND256(d,e,f,g,h,a,b,c);
		ROUND256(c,d,e,f,g,h,a,b);
		ROUND256(b,c,d,e,f,g,h,a);
	} while (j < 64);

	/* Compute the current intermediate hash value */
	context->state[0] += a;
	context->state[1] += b;
	context->state[2] += c;
	context->state[3] += d;
	context->state[4] += e;
	context->state[5] += f;
	context->state[6] += g;
	context->state[7] += h;

	/* Clean up */
	a = b = c = d = e = f = g = h = T1 = 0;
}