08fca7a56b
for counter mode), and AES-GCM. Both of these modes have been added to the aesni module. Included is a set of tests to validate that the software and aesni module calculate the correct values. These use the NIST KAT test vectors. To run the test, you will need to install a soon to be committed port, nist-kat that will install the vectors. Using a port is necessary as the test vectors are around 25MB. All the man pages were updated. I have added a new man page, crypto.7, which includes a description of how to use each mode. All the new modes and some other AES modes are present. It would be good for someone else to go through and document the other modes. A new ioctl was added to support AEAD modes which AES-GCM is one of them. Without this ioctl, it is not possible to test AEAD modes from userland. Add a timing safe bcmp for use to compare MACs. Previously we were using bcmp which could leak timing info and result in the ability to forge messages. Add a minor optimization to the aesni module so that single segment mbufs don't get copied and instead are updated in place. The aesni module needs to be updated to support blocked IO so segmented mbufs don't have to be copied. We require that the IV be specified for all calls for both GCM and ICM. This is to ensure proper use of these functions. Obtained from: p4: //depot/projects/opencrypto Relnotes: yes Sponsored by: FreeBSD Foundation Sponsored by: NetGate
129 lines
3.9 KiB
C
129 lines
3.9 KiB
C
/*-
|
|
* Copyright (c) 2014 The FreeBSD Foundation
|
|
* All rights reserved.
|
|
*
|
|
* This software was developed by John-Mark Gurney under
|
|
* the sponsorship of the FreeBSD Foundation and
|
|
* Rubicon Communications, LLC (Netgate).
|
|
* Redistribution and use in source and binary forms, with or without
|
|
* modification, are permitted provided that the following conditions
|
|
* are met:
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
* notice, this list of conditions and the following disclaimer.
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
* documentation and/or other materials provided with the distribution.
|
|
*
|
|
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
* SUCH DAMAGE.
|
|
*
|
|
* $FreeBSD$
|
|
*
|
|
*/
|
|
|
|
#ifndef _GFMULT_H_
|
|
#define _GFMULT_H_
|
|
|
|
#ifdef __APPLE__
|
|
#define __aligned(x) __attribute__((__aligned__(x)))
|
|
#define be64dec(buf) __builtin_bswap64(*(uint64_t *)buf)
|
|
#define be64enc(buf, x) (*(uint64_t *)buf = __builtin_bswap64(x))
|
|
#else
|
|
#include <sys/endian.h>
|
|
#endif
|
|
|
|
#ifdef _KERNEL
|
|
#include <sys/types.h>
|
|
#else
|
|
#include <stdint.h>
|
|
#include <strings.h>
|
|
#endif
|
|
|
|
#define REQ_ALIGN (16 * 4)
|
|
/*
|
|
* The rows are striped across cache lines. Note that the indexes
|
|
* are bit reversed to make accesses quicker.
|
|
*/
|
|
struct gf128table {
|
|
uint32_t a[16] __aligned(REQ_ALIGN); /* bits 0 - 31 */
|
|
uint32_t b[16] __aligned(REQ_ALIGN); /* bits 63 - 32 */
|
|
uint32_t c[16] __aligned(REQ_ALIGN); /* bits 95 - 64 */
|
|
uint32_t d[16] __aligned(REQ_ALIGN); /* bits 127 - 96 */
|
|
} __aligned(REQ_ALIGN);
|
|
|
|
/*
|
|
* A set of tables that contain h, h^2, h^3, h^4. To be used w/ gf128_mul4.
|
|
*/
|
|
struct gf128table4 {
|
|
struct gf128table tbls[4];
|
|
};
|
|
|
|
/*
|
|
* GCM per spec is bit reversed in memory. So byte 0 is really bit reversed
|
|
* and contains bits 0-7. We can deal w/ this by using right shifts and
|
|
* related math instead of having to bit reverse everything. This means that
|
|
* the low bits are in v[0] (bits 0-63) and reverse order, while the high
|
|
* bits are in v[1] (bits 64-127) and reverse order. The high bit of v[0] is
|
|
* bit 0, and the low bit of v[1] is bit 127.
|
|
*/
|
|
struct gf128 {
|
|
uint64_t v[2];
|
|
};
|
|
|
|
/* Note that we don't bit reverse in MAKE_GF128. */
|
|
#define MAKE_GF128(a, b) ((struct gf128){.v = { (a), (b) } })
|
|
#define GF128_EQ(a, b) ((((a).v[0] ^ (b).v[0]) | \
|
|
((a).v[1] ^ (b).v[1])) == 0)
|
|
|
|
static inline struct gf128
|
|
gf128_read(const uint8_t *buf)
|
|
{
|
|
struct gf128 r;
|
|
|
|
r.v[0] = be64dec(buf);
|
|
buf += sizeof(uint64_t);
|
|
|
|
r.v[1] = be64dec(buf);
|
|
|
|
return r;
|
|
}
|
|
|
|
static inline void
|
|
gf128_write(struct gf128 v, uint8_t *buf)
|
|
{
|
|
uint64_t tmp;
|
|
|
|
be64enc(buf, v.v[0]);
|
|
buf += sizeof tmp;
|
|
|
|
be64enc(buf, v.v[1]);
|
|
}
|
|
|
|
static inline struct gf128 __pure /* XXX - __pure2 instead */
|
|
gf128_add(struct gf128 a, struct gf128 b)
|
|
{
|
|
a.v[0] ^= b.v[0];
|
|
a.v[1] ^= b.v[1];
|
|
|
|
return a;
|
|
}
|
|
|
|
void gf128_genmultable(struct gf128 h, struct gf128table *t);
|
|
void gf128_genmultable4(struct gf128 h, struct gf128table4 *t);
|
|
struct gf128 gf128_mul(struct gf128 v, struct gf128table *tbl);
|
|
struct gf128 gf128_mul4(struct gf128 a, struct gf128 b, struct gf128 c,
|
|
struct gf128 d, struct gf128table4 *tbl);
|
|
struct gf128 gf128_mul4b(struct gf128 r, const uint8_t *v,
|
|
struct gf128table4 *tbl);
|
|
|
|
#endif /* _GFMULT_H_ */
|