shithub: riscv

Download patch

ref: 7effba9d98e8a39ee1504192d02d40e86d90b19a
parent: a3c155704117d99f4c7486f9cf48d0a281997ff1
author: cinap_lenrek <[email protected]>
date: Thu Oct 22 03:17:25 EDT 2015

libsec: add poly1305

--- a/sys/include/libsec.h
+++ b/sys/include/libsec.h
@@ -141,6 +141,7 @@
 	SHA2_512dlen=	64,	/* SHA-512 digest length */
 	MD4dlen=	16,	/* MD4 digest length */
 	MD5dlen=	16,	/* MD5 digest length */
+	Poly1305dlen=	16,	/* Poly1305 digest length */
 
 	Hmacblksz	= 64,	/* in bytes; from rfc2104 */
 };
@@ -188,6 +189,8 @@
 MD5state*	md5unpickle(char*);
 char*		sha1pickle(SHA1state*);
 SHA1state*	sha1unpickle(char*);
+
+DigestState*	poly1305(uchar*, ulong, uchar*, ulong, uchar*, DigestState*);
 
 /*
  * random number generation
--- a/sys/man/2/sechash
+++ b/sys/man/2/sechash
@@ -2,7 +2,7 @@
 .SH NAME
 md4, md5,
 sha1, sha2_224, sha2_256, sha2_384, sha2_512,
-ripemd160,
+ripemd160, poly1305,
 hmac_x, hmac_md5,
 hmac_sha1, hmac_sha2_224, hmac_sha2_256, hmac_sha2_384, hmac_sha2_512,
 md5pickle, md5unpickle,
@@ -58,6 +58,8 @@
 .Ti
 DS*	ripemd160(uchar *data, ulong dlen, uchar *digest, DS *state)
 .Ti
+DS*	poly1305(uchar *p, ulong len, uchar *key, ulong klen, uchar *digest, DS *state)
+.Ti
 DS*	hmac_x(uchar *p, ulong len, uchar *key, ulong klen, uchar *digest, DS *s, DS*(*x)(uchar*, ulong, uchar*, DS*), int xlen)
 .Ti
 DS*	hmac_md5(uchar *data, ulong dlen, uchar *key, ulong klen, uchar *digest, DS *state)
@@ -92,6 +94,7 @@
 .IR sha2_384 ,
 .IR sha2_512 ,
 .IR ripemd160 ,
+.IR poly1305 ,
 .IR hmac_md5 ,
 .IR hmac_sha1 ,
 .IR hmac_sha2_224 ,
@@ -148,6 +151,10 @@
 internally, but
 .I hmac_x
 is not intended for general use.
+.PP
+.IR Poly1305
+is a one-time authenticator designed by D. J. Bernstein. It takes a 32-byte
+one-time key and a message and produces a 16-byte tag.
 .PP
 The functions
 .I md5pickle
--- a/sys/src/libsec/port/mkfile
+++ b/sys/src/libsec/port/mkfile
@@ -7,6 +7,7 @@
 	hmac.c md5.c md5block.c md4.c sha1.c sha1block.c\
 	sha2_64.c sha2_128.c sha2block64.c sha2block128.c\
 	sha1pickle.c md5pickle.c\
+	poly1305.c\
 	rc4.c\
 	genrandom.c prng.c fastrand.c nfastrand.c\
 	probably_prime.c smallprimetest.c genprime.c dsaprimes.c\
--- /dev/null
+++ b/sys/src/libsec/port/poly1305.c
@@ -1,0 +1,192 @@
+#include <u.h>
+#include <libc.h>
+#include <libsec.h>
+
+/*
+	poly1305 implementation using 32 bit * 32 bit = 64 bit multiplication and 64 bit addition
+
+	derived from http://github.com/floodberry/poly1305-donna
+*/
+
+#define U8TO32(p)	((u32int)(p)[0] | (u32int)(p)[1]<<8 | (u32int)(p)[2]<<16 | (u32int)(p)[3]<<24)
+#define U32TO8(p, v)	(p)[0]=(v), (p)[1]=(v)>>8, (p)[2]=(v)>>16, (p)[3]=(v)>>24
+
+/* (r,s) = (key[0:15],key[16:31]), the one time key */
+DigestState*
+poly1305(uchar *m, ulong len, uchar *key, ulong klen, uchar *digest, DigestState *s)
+{
+	u32int r0,r1,r2,r3,r4, s1,s2,s3,s4, h0,h1,h2,h3,h4, g0,g1,g2,g3,g4;
+	u64int d0,d1,d2,d3,d4, f;
+	u32int hibit, mask, c;
+
+	if(s == nil){
+		s = malloc(sizeof(*s));
+		if(s == nil)
+			return nil;
+		memset(s, 0, sizeof(*s));
+		s->malloced = 1;
+	}
+
+	if(s->seeded == 0){
+		assert(klen == 32);
+
+		/* r &= 0xffffffc0ffffffc0ffffffc0fffffff */
+		s->state[0] = (U8TO32(&key[ 0])     ) & 0x3ffffff;
+		s->state[1] = (U8TO32(&key[ 3]) >> 2) & 0x3ffff03;
+		s->state[2] = (U8TO32(&key[ 6]) >> 4) & 0x3ffc0ff;
+		s->state[3] = (U8TO32(&key[ 9]) >> 6) & 0x3f03fff;
+		s->state[4] = (U8TO32(&key[12]) >> 8) & 0x00fffff;
+
+		/* h = 0 */
+		s->state[5] = 0;
+		s->state[6] = 0;
+		s->state[7] = 0;
+		s->state[8] = 0;
+		s->state[9] = 0;
+
+		/* save pad for later */
+		s->state[10] = U8TO32(&key[16]);
+		s->state[11] = U8TO32(&key[20]);
+		s->state[12] = U8TO32(&key[24]);
+		s->state[13] = U8TO32(&key[28]);
+
+		s->seeded = 1;
+	}
+
+	if(s->blen){
+		c = 16 - s->blen;
+		if(c > len)
+			c = len;
+		memmove(s->buf + s->blen, m, c);
+		len -= c, m += c;
+		s->blen += c;
+		if(s->blen == 16){
+			s->blen = 0;
+			poly1305(s->buf, 16, key, klen, nil, s);
+		}
+	}
+
+	r0 = s->state[0];
+	r1 = s->state[1];
+	r2 = s->state[2];
+	r3 = s->state[3];
+	r4 = s->state[4];
+
+	h0 = s->state[5];
+	h1 = s->state[6];
+	h2 = s->state[7];
+	h3 = s->state[8];
+	h4 = s->state[9];
+
+	s1 = r1 * 5;
+	s2 = r2 * 5;
+	s3 = r3 * 5;
+	s4 = r4 * 5;
+
+	hibit = 1<<24;	/* 1<<128 */
+
+	while(len >= 16){
+Block:
+		/* h += m[i] */
+		h0 += (U8TO32(&m[0])     ) & 0x3ffffff;
+		h1 += (U8TO32(&m[3]) >> 2) & 0x3ffffff;
+		h2 += (U8TO32(&m[6]) >> 4) & 0x3ffffff;
+		h3 += (U8TO32(&m[9]) >> 6) & 0x3ffffff;
+		h4 += (U8TO32(&m[12])>> 8) | hibit;
+
+		/* h *= r */
+		d0 = ((u64int)h0 * r0) + ((u64int)h1 * s4) + ((u64int)h2 * s3) + ((u64int)h3 * s2) + ((u64int)h4 * s1);
+		d1 = ((u64int)h0 * r1) + ((u64int)h1 * r0) + ((u64int)h2 * s4) + ((u64int)h3 * s3) + ((u64int)h4 * s2);
+		d2 = ((u64int)h0 * r2) + ((u64int)h1 * r1) + ((u64int)h2 * r0) + ((u64int)h3 * s4) + ((u64int)h4 * s3);
+		d3 = ((u64int)h0 * r3) + ((u64int)h1 * r2) + ((u64int)h2 * r1) + ((u64int)h3 * r0) + ((u64int)h4 * s4);
+		d4 = ((u64int)h0 * r4) + ((u64int)h1 * r3) + ((u64int)h2 * r2) + ((u64int)h3 * r1) + ((u64int)h4 * r0);
+
+		/* (partial) h %= p */
+		              c = (u32int)(d0 >> 26); h0 = (u32int)d0 & 0x3ffffff;
+		d1 += c;      c = (u32int)(d1 >> 26); h1 = (u32int)d1 & 0x3ffffff;
+		d2 += c;      c = (u32int)(d2 >> 26); h2 = (u32int)d2 & 0x3ffffff;
+		d3 += c;      c = (u32int)(d3 >> 26); h3 = (u32int)d3 & 0x3ffffff;
+		d4 += c;      c = (u32int)(d4 >> 26); h4 = (u32int)d4 & 0x3ffffff;
+		h0 += c * 5;  c = (h0 >> 26); h0 = h0 & 0x3ffffff;
+		h1 += c;
+
+		len -= 16, m += 16;
+	}
+
+	if(len){
+		s->blen = len;
+		memmove(s->buf, m, len);
+	}
+
+	if(digest == nil){
+		s->state[5] = h0;
+		s->state[6] = h1;
+		s->state[7] = h2;
+		s->state[8] = h3;
+		s->state[9] = h4;
+		return s;
+	}
+
+	if(len){
+		m = s->buf;
+		m[len++] = 1;
+		while(len < 16)
+			m[len++] = 0;
+		hibit = 0;
+		goto Block;
+	}
+
+	             c = h1 >> 26; h1 = h1 & 0x3ffffff;
+	h2 +=     c; c = h2 >> 26; h2 = h2 & 0x3ffffff;
+	h3 +=     c; c = h3 >> 26; h3 = h3 & 0x3ffffff;
+	h4 +=     c; c = h4 >> 26; h4 = h4 & 0x3ffffff;
+	h0 += c * 5; c = h0 >> 26; h0 = h0 & 0x3ffffff;
+	h1 +=     c;
+
+	/* compute h + -p */
+	g0 = h0 + 5; c = g0 >> 26; g0 &= 0x3ffffff;
+	g1 = h1 + c; c = g1 >> 26; g1 &= 0x3ffffff;
+	g2 = h2 + c; c = g2 >> 26; g2 &= 0x3ffffff;
+	g3 = h3 + c; c = g3 >> 26; g3 &= 0x3ffffff;
+	g4 = h4 + c - (1 << 26);
+
+	/* select h if h < p, or h + -p if h >= p */
+	mask = (g4 >> 31) - 1;
+	g0 &= mask;
+	g1 &= mask;
+	g2 &= mask;
+	g3 &= mask;
+	g4 &= mask;
+	mask = ~mask;
+	h0 = (h0 & mask) | g0;
+	h1 = (h1 & mask) | g1;
+	h2 = (h2 & mask) | g2;
+	h3 = (h3 & mask) | g3;
+	h4 = (h4 & mask) | g4;
+
+	/* h = h % (2^128) */
+	h0 = (h0      ) | (h1 << 26);
+	h1 = (h1 >>  6) | (h2 << 20);
+	h2 = (h2 >> 12) | (h3 << 14);
+	h3 = (h3 >> 18) | (h4 <<  8);
+	
+	/* digest = (h + pad) % (2^128) */
+	f = (u64int)h0 + s->state[10]            ; h0 = (u32int)f;
+	f = (u64int)h1 + s->state[11] + (f >> 32); h1 = (u32int)f;
+	f = (u64int)h2 + s->state[12] + (f >> 32); h2 = (u32int)f;
+	f = (u64int)h3 + s->state[13] + (f >> 32); h3 = (u32int)f;
+
+	U32TO8(&digest[0], h0);
+	U32TO8(&digest[4], h1);
+	U32TO8(&digest[8], h2);
+	U32TO8(&digest[12], h3);
+
+	if(s->malloced){
+		memset(s, 0, sizeof(*s));
+		free(s);
+		return nil;
+	}
+
+	memset(s, 0, sizeof(*s));
+	return nil;
+}