freebsd-dev/sys/contrib/libb2/blake2sp.c
Conrad Meyer 0e33efe4e4 Import Blake2 algorithms (blake2b, blake2s) from libb2
The upstream repository is on github BLAKE2/libb2.  Files landed in
sys/contrib/libb2 are the unmodified upstream files, except for one
difference:  secure_zero_memory's contents have been replaced with
explicit_bzero() only because the previous implementation broke powerpc
link.  Preferential use of explicit_bzero() is in progress upstream, so
it is anticipated we will be able to drop this diff in the future.

sys/crypto/blake2 contains the source files needed to port libb2 to our
build system, a wrapped (limited) variant of the algorithm to match the API
of our auth_transform softcrypto abstraction, incorporation into the Open
Crypto Framework (OCF) cryptosoft(4) driver, as well as an x86 SSE/AVX
accelerated OCF driver, blake2(4).

Optimized variants of blake2 are compiled for a number of x86 machines
(anything from SSE2 to AVX + XOP).  On those machines, FPU context will need
to be explicitly saved before using blake2(4)-provided algorithms directly.
Use via cryptodev / OCF saves FPU state automatically, and use via the
auth_transform softcrypto abstraction does not use FPU.

The intent of the OCF driver is mostly to enable testing in userspace via
/dev/crypto.  ATF tests are added with published KAT test vectors to
validate correctness.

Reviewed by:	jhb, markj
Obtained from:	github BLAKE2/libb2
Differential Revision:	https://reviews.freebsd.org/D14662
2018-03-21 16:18:14 +00:00

275 lines
7.3 KiB
C

/*
BLAKE2 reference source code package - optimized C implementations
Written in 2012 by Samuel Neves <sneves@dei.uc.pt>
To the extent possible under law, the author(s) have dedicated all copyright
and related and neighboring rights to this software to the public domain
worldwide. This software is distributed without any warranty.
You should have received a copy of the CC0 Public Domain Dedication along with
this software. If not, see <http://creativecommons.org/publicdomain/zero/1.0/>.
*/
#include <stdlib.h>
#include <string.h>
#include <stdio.h>
#if defined(_OPENMP)
#include <omp.h>
#endif
#include "blake2.h"
#include "blake2-impl.h"
#define PARALLELISM_DEGREE 8
static int blake2sp_init_leaf( blake2s_state *S, uint8_t outlen, uint8_t keylen, uint64_t offset )
{
blake2s_param P[1];
P->digest_length = outlen;
P->key_length = keylen;
P->fanout = PARALLELISM_DEGREE;
P->depth = 2;
P->leaf_length = 0;
store48( P->node_offset, offset );
P->node_depth = 0;
P->inner_length = BLAKE2S_OUTBYTES;
memset( P->salt, 0, sizeof( P->salt ) );
memset( P->personal, 0, sizeof( P->personal ) );
blake2s_init_param( S, P );
S->outlen = P->inner_length;
return 0;
}
static int blake2sp_init_root( blake2s_state *S, uint8_t outlen, uint8_t keylen )
{
blake2s_param P[1];
P->digest_length = outlen;
P->key_length = keylen;
P->fanout = PARALLELISM_DEGREE;
P->depth = 2;
P->leaf_length = 0;
store48( P->node_offset, 0ULL );
P->node_depth = 1;
P->inner_length = BLAKE2S_OUTBYTES;
memset( P->salt, 0, sizeof( P->salt ) );
memset( P->personal, 0, sizeof( P->personal ) );
blake2s_init_param( S, P );
S->outlen = P->digest_length;
return 0;
}
int blake2sp_init( blake2sp_state *S, size_t outlen )
{
if( !outlen || outlen > BLAKE2S_OUTBYTES ) return -1;
memset( S->buf, 0, sizeof( S->buf ) );
S->buflen = 0;
if( blake2sp_init_root( S->R, ( uint8_t ) outlen, 0 ) < 0 )
return -1;
for( size_t i = 0; i < PARALLELISM_DEGREE; ++i )
if( blake2sp_init_leaf( S->S[i], ( uint8_t ) outlen, 0, i ) < 0 ) return -1;
S->R->last_node = 1;
S->S[PARALLELISM_DEGREE - 1]->last_node = 1;
S->outlen = ( uint8_t ) outlen;
return 0;
}
int blake2sp_init_key( blake2sp_state *S, size_t outlen, const void *key, size_t keylen )
{
if( !outlen || outlen > BLAKE2S_OUTBYTES ) return -1;
if( !key || !keylen || keylen > BLAKE2S_KEYBYTES ) return -1;
memset( S->buf, 0, sizeof( S->buf ) );
S->buflen = 0;
if( blake2sp_init_root( S->R, ( uint8_t ) outlen, ( uint8_t ) keylen ) < 0 )
return -1;
for( size_t i = 0; i < PARALLELISM_DEGREE; ++i )
if( blake2sp_init_leaf( S->S[i], ( uint8_t ) outlen, ( uint8_t ) keylen, i ) < 0 )
return -1;
S->R->last_node = 1;
S->S[PARALLELISM_DEGREE - 1]->last_node = 1;
S->outlen = ( uint8_t ) outlen;
{
uint8_t block[BLAKE2S_BLOCKBYTES];
memset( block, 0, BLAKE2S_BLOCKBYTES );
memcpy( block, key, keylen );
for( size_t i = 0; i < PARALLELISM_DEGREE; ++i )
blake2s_update( S->S[i], block, BLAKE2S_BLOCKBYTES );
secure_zero_memory( block, BLAKE2S_BLOCKBYTES ); /* Burn the key from stack */
}
return 0;
}
int blake2sp_update( blake2sp_state *S, const uint8_t *in, size_t inlen )
{
size_t left = S->buflen;
size_t fill = sizeof( S->buf ) - left;
if( left && inlen >= fill )
{
memcpy( S->buf + left, in, fill );
for( size_t i = 0; i < PARALLELISM_DEGREE; ++i )
blake2s_update( S->S[i], S->buf + i * BLAKE2S_BLOCKBYTES, BLAKE2S_BLOCKBYTES );
in += fill;
inlen -= fill;
left = 0;
}
#if defined(_OPENMP)
omp_set_num_threads(PARALLELISM_DEGREE);
#pragma omp parallel shared(S)
#else
for( size_t id__ = 0; id__ < PARALLELISM_DEGREE; ++id__ )
#endif
{
#if defined(_OPENMP)
size_t id__ = ( size_t ) omp_get_thread_num();
#endif
size_t inlen__ = inlen;
const uint8_t *in__ = ( const uint8_t * )in;
in__ += id__ * BLAKE2S_BLOCKBYTES;
while( inlen__ >= PARALLELISM_DEGREE * BLAKE2S_BLOCKBYTES )
{
blake2s_update( S->S[id__], in__, BLAKE2S_BLOCKBYTES );
in__ += PARALLELISM_DEGREE * BLAKE2S_BLOCKBYTES;
inlen__ -= PARALLELISM_DEGREE * BLAKE2S_BLOCKBYTES;
}
}
in += inlen - inlen % ( PARALLELISM_DEGREE * BLAKE2S_BLOCKBYTES );
inlen %= PARALLELISM_DEGREE * BLAKE2S_BLOCKBYTES;
if( inlen > 0 )
memcpy( S->buf + left, in, inlen );
S->buflen = ( uint32_t ) left + ( uint32_t ) inlen;
return 0;
}
int blake2sp_final( blake2sp_state *S, uint8_t *out, size_t outlen )
{
uint8_t hash[PARALLELISM_DEGREE][BLAKE2S_OUTBYTES];
if(S->outlen != outlen) return -1;
for( size_t i = 0; i < PARALLELISM_DEGREE; ++i )
{
if( S->buflen > i * BLAKE2S_BLOCKBYTES )
{
size_t left = S->buflen - i * BLAKE2S_BLOCKBYTES;
if( left > BLAKE2S_BLOCKBYTES ) left = BLAKE2S_BLOCKBYTES;
blake2s_update( S->S[i], S->buf + i * BLAKE2S_BLOCKBYTES, left );
}
blake2s_final( S->S[i], hash[i], BLAKE2S_OUTBYTES );
}
for( size_t i = 0; i < PARALLELISM_DEGREE; ++i )
blake2s_update( S->R, hash[i], BLAKE2S_OUTBYTES );
blake2s_final( S->R, out, outlen );
return 0;
}
int blake2sp( uint8_t *out, const void *in, const void *key, size_t outlen, size_t inlen, size_t keylen )
{
uint8_t hash[PARALLELISM_DEGREE][BLAKE2S_OUTBYTES];
blake2s_state S[PARALLELISM_DEGREE][1];
blake2s_state FS[1];
/* Verify parameters */
if ( NULL == in && inlen > 0 ) return -1;
if ( NULL == out ) return -1;
if ( NULL == key && keylen > 0 ) return -1;
if( !outlen || outlen > BLAKE2S_OUTBYTES ) return -1;
if( keylen > BLAKE2S_KEYBYTES ) return -1;
for( size_t i = 0; i < PARALLELISM_DEGREE; ++i )
if( blake2sp_init_leaf( S[i], ( uint8_t ) outlen, ( uint8_t ) keylen, i ) < 0 )
return -1;
S[PARALLELISM_DEGREE - 1]->last_node = 1; // mark last node
if( keylen > 0 )
{
uint8_t block[BLAKE2S_BLOCKBYTES];
memset( block, 0, BLAKE2S_BLOCKBYTES );
memcpy( block, key, keylen );
for( size_t i = 0; i < PARALLELISM_DEGREE; ++i )
blake2s_update( S[i], block, BLAKE2S_BLOCKBYTES );
secure_zero_memory( block, BLAKE2S_BLOCKBYTES ); /* Burn the key from stack */
}
#if defined(_OPENMP)
omp_set_num_threads(PARALLELISM_DEGREE);
#pragma omp parallel shared(S,hash)
#else
for( size_t id__ = 0; id__ < PARALLELISM_DEGREE; ++id__ )
#endif
{
#if defined(_OPENMP)
size_t id__ = ( size_t ) omp_get_thread_num();
#endif
size_t inlen__ = inlen;
const uint8_t *in__ = ( const uint8_t * )in;
in__ += id__ * BLAKE2S_BLOCKBYTES;
while( inlen__ >= PARALLELISM_DEGREE * BLAKE2S_BLOCKBYTES )
{
blake2s_update( S[id__], in__, BLAKE2S_BLOCKBYTES );
in__ += PARALLELISM_DEGREE * BLAKE2S_BLOCKBYTES;
inlen__ -= PARALLELISM_DEGREE * BLAKE2S_BLOCKBYTES;
}
if( inlen__ > id__ * BLAKE2S_BLOCKBYTES )
{
const size_t left = inlen__ - id__ * BLAKE2S_BLOCKBYTES;
const size_t len = left <= BLAKE2S_BLOCKBYTES ? left : BLAKE2S_BLOCKBYTES;
blake2s_update( S[id__], in__, len );
}
blake2s_final( S[id__], hash[id__], BLAKE2S_OUTBYTES );
}
if( blake2sp_init_root( FS, ( uint8_t ) outlen, ( uint8_t ) keylen ) < 0 )
return -1;
FS->last_node = 1;
for( size_t i = 0; i < PARALLELISM_DEGREE; ++i )
blake2s_update( FS, hash[i], BLAKE2S_OUTBYTES );
return blake2s_final( FS, out, outlen );
}