The Nettle low-level cryptographic library.

Table of Contents

Node:Top, Next:, Previous:(dir), Up:(dir)

Nettle

This document describes the nettle low-level cryptographic library. You can use the library directly from your C programs, or (recommended) write or use an object-oriented wrapper for your favorite language or application.

This manual corresponds to version 1.5 of the library.

Node:Introduction, Next:, Previous:Top, Up:Top

Introduction

Nettle is a cryptographic library that is designed to fit easily in more or less any context: In crypto toolkits for object-oriented languages (C++, Python, Pike, ...), in applications like LSH or GNUPG, or even in kernel space. In most contexts, you need more than the basic cryptographic algorithms, you also need some way to keep track of available algorithms, their properties and variants. You often have some algorithm selection process, often dictated by a protocol you want to implement.

And as the requirements of applications differ in subtle and not so subtle ways, an API that fits one application well can be a pain to use in a different context. And that is why there are so many different cryptographic libraries around.

Nettle tries to avoid this problem by doing one thing, the low-level crypto stuff, and providing a simple but general interface to it. In particular, Nettle doesn't do algorithm selection. It doesn't do memory allocation. It doesn't do any I/O.

The idea is that one can build several application and context specific interfaces on top of Nettle, and share the code, test cases, benchmarks, documentation, etc. For this first version, the only application using Nettle is LSH, and it uses an object-oriented abstraction on top of the library.

This manual explains how to use the Nettle library. It also tries to provide some background on the cryptography, and advice on how to best put it to use.

Node:Copyright, Next:, Previous:Introduction, Up:Top

Copyright

Nettle is distributed under the GNU General Public License (GPL) (see the file COPYING for details). However, most of the individual files are dual licensed under less restrictive licenses like the GNU Lesser General Public License (LGPL), or are in the public domain. This means that if you don't use the parts of nettle that are GPL-only, you have the option to use the Nettle library just as if it were licensed under the LGPL. To find the current status of particular files, you have to read the copyright notices at the top of the files.

A list of the supported algorithms, their origins and licenses:

AES
The implementation of the AES cipher (also known as rijndael) is written by Rafael Sevilla. Released under the LGPL.
ARCFOUR
The implementation of the ARCFOUR (also known as RC4) cipher is written by Niels Möller. Released under the LGPL.
BLOWFISH
The implementation of the BLOWFISH cipher is written by Werner Koch, copyright owned by the Free Software Foundation. Also hacked by Ray Dassen and Niels Möller. Released under the GPL.
CAST128
The implementation of the CAST128 cipher is written by Steve Reid. Released into the public domain.
DES
The implementation of the DES cipher is written by Dana L. How, and released under the LGPL.
MD5
The implementation of the MD5 message digest is written by Colin Plumb. It has been hacked some more by Andrew Kuchling and Niels Möller. Released into the public domain.
SERPENT
The implementation of the SERPENT cipher is written by Ross Anderson, Eli Biham, and Lars Knudsen, adapted to LSH by Rafael Sevilla, and to Nettle by Niels Möller. Released under the GPL.
SHA1
The implementation of the SHA1 message digest is written by Peter Gutmann, and hacked some more by Andrew Kuchling and Niels Möller. Released into the public domain.
TWOFISH
The implementation of the TWOFISH cipher is written by Ruud de Rooij. Released under the LGPL.

Node:Conventions, Next:, Previous:Copyright, Up:Top

Conventions

For each supported algorithm, there is an include file that defines a context struct, a few constants, and declares functions for operating on the context. The context struct encapsulates all information needed by the algorithm, and it can be copied or moved in memory with no unexpected effects.

For consistency, functions for different algorithms are very similar, but there are some differences, for instance reflecting if the key setup or encryption function differ for encryption and encryption, and whether or not key setup can fail. There are also differences between algorithms that don't show in function prototypes, but which the application must nevertheless be aware of. There is no big difference between the functions for stream ciphers and for block ciphers, although they should be used quite differently by the application.

If your application uses more than one algorithm, you should probably create an interface that is tailor-made for your needs, and then write a few lines of glue code on top of Nettle.

By convention, for an algorithm named foo, the struct tag for the context struct is foo_ctx, constants and functions uses prefixes like FOO_BLOCK_SIZE (a constant) and foo_set_key (a function).

In all functions, strings are represented with an explicit length, of type unsigned, and a pointer of type uint8_t * or const uint8_t *. For functions that transform one string to another, the argument order is length, destination pointer and source pointer. Source and destination areas are of the same length. Source and destination may be the same, so that you can process strings in place, but they must not overlap in any other way.

Node:Example, Next:, Previous:Conventions, Up:Top

Example

A simple example program that reads a file from standard in and writes its SHA1 checksum on standard output should give the flavor of Nettle. 

/* FIXME: This code is untested. */
#include <stdio.h>
#include <stdlib.h>

#include <nettle/sha.h>

#define BUF_SIZE 1000

static void
display_hex(unsigned length, uint8_t *data)
{
  static const char digits[16] = "0123456789abcdef";
  unsigned i;

  for (i = 0; i<length; i++)
  {
    uint8_t byte = data[i];
    printf("%c%c ", digits[(byte / 16) & 0xf], digits[byte & 0xf]);
  }
}

int
main(int argc, char **argv)
{
  struct sha1_ctx ctx;
  uint8_t buffer[BUF_SIZE];
  uint8_t digest[SHA1_DIGEST_SIZE];

  sha1_init(&ctx);
  for (;;)
  {
    int done = fread(buffer, 1, sizeof(buffer), stdin);
    if (done <= 0)
      break;
    sha1_update(&ctx, done, buf);
  }
  if (ferror(stdin))
    return EXIT_FAILURE;

  sha1_digest(&ctx, SHA1_DIGEST_SIZE, digest);

  display_hex(SHA1_DIGEST_SIZE, digest);
  return EXIT_SUCCESS;
}

Node:Reference, Next:, Previous:Example, Up:Top

Reference

This chapter describes all the Nettle functions, grouped by family.

Node:Hash functions, Next:, Previous:Reference, Up:Reference

Hash functions

A cryptographic hash function is a function that takes variable size strings, and maps them to strings of fixed, short, length. There are naturally lots of collisions, as there are more possible 1MB files than 20 byte strings. But the function is constructed such that is hard to find the collisions. More precisely, a cryptographic hash function H should have the following properties:


One-way
Given a hash value H(x) it is hard to find a string x that hashes to that value.
Collision-resistant
It is hard to find two different strings, x and y, such that H(x) = H(y).

Hash functions are useful as building blocks for digital signatures, message authentication codes, pseudo random generators, association of unique id:s to documents, and many other things.

MD5

MD5 is a message digest function constructed by Ronald Rivest, and described in RFC 1321. It outputs message digests of 128 bits, or 16 octets. Nettle defines MD5 in <nettle/md5.h>.

struct md5_ctx Context struct

MD5_DIGEST_SIZE Constant
The size of an MD5 digest, i.e. 16.

MD5_DATA_SIZE Constant
The internal block size of MD5. Useful for some special constructions, in particular HMAC-MD5.

void md5_init (struct md5_ctx *ctx) Function
Initialize the MD5 state.

void md5_update (struct md5_ctx *ctx, unsigned length, const uint8_t *data) Function
Hash some more data.

void md5_digest (struct md5_ctx *ctx, unsigned length, uint8_t *digest) Function
Performs final processing and extracts the message digest, writing it to digest. length may be smaller than MD5_DIGEST_SIZE, in which case only the first length octets of the digest are written.

This function also resets the context in the same way as md5_init.

The normal way to use MD5 is to call the functions in order: First md5_init, then md5_update zero or more times, and finally md5_digest. After md5_digest, the context is reset to its initial state, so you can start over calling md5_update to hash new data.

To start over, you can call md5_init at any time.

SHA1

SHA1 is a hash function specified by NIST (The U.S. National Institute for Standards and Technology). It outputs hash values of 160 bits, or 20 octets. Nettle defines SHA1 in <nettle/sha.h>.

The functions are analogous to the MD5 ones.

struct sha1_ctx Context struct

SHA1_DIGEST_SIZE Constant
The size of an SHA1 digest, i.e. 20.

SHA1_DATA_SIZE Constant
The internal block size of SHA1. Useful for some special constructions, in particular HMAC-SHA1.

void sha1_init (struct sha1_ctx *ctx) Function
Initialize the SHA1 state.

void sha1_update (struct sha1_ctx *ctx, unsigned length, const uint8_t *data) Function
Hash some more data.

void sha1_digest (struct sha1_ctx *ctx, unsigned length, uint8_t *digest) Function
Performs final processing and extracts the message digest, writing it to digest. length may be smaller than SHA1_DIGEST_SIZE, in which case only the first length octets of the digest are written.

This function also resets the context in the same way as sha1_init.

SHA256

SHA256 is another hash function specified by NIST, intended as a replacement for SHA1, generating larger digests. It outputs hash values of 256 bits, or 32 octets. Nettle defines SHA256 in <nettle/sha.h>.

The functions are analogous to the MD5 ones.

struct sha256_ctx Context struct

SHA256_DIGEST_SIZE Constant
The size of an SHA256 digest, i.e. 20.

SHA256_DATA_SIZE Constant
The internal block size of SHA256. Useful for some special constructions, in particular HMAC-SHA256.

void sha256_init (struct sha256_ctx *ctx) Function
Initialize the SHA256 state.

void sha256_update (struct sha256_ctx *ctx, unsigned length, const uint8_t *data) Function
Hash some more data.

void sha256_digest (struct sha256_ctx *ctx, unsigned length, uint8_t *digest) Function
Performs final processing and extracts the message digest, writing it to digest. length may be smaller than SHA256_DIGEST_SIZE, in which case only the first length octets of the digest are written.

This function also resets the context in the same way as sha256_init.

struct nettle_hash

Nettle includes a struct including information about the supported hash functions. It is defined in <nettle/nettle-meta.h>, and is used by Nettle's implementation of HMAC see Keyed hash functions.

struct nettle_hash name context_size digest_size block_size init update digest Meta struct
The last three attributes are function pointers, of types nettle_hash_init_func, nettle_hash_update_func, and nettle_hash_digest_func. The first argument to these functions is void * pointer so a context struct, which is of size context_size.

struct nettle_cipher nettle_md5 Constant Struct
struct nettle_cipher nettle_sha1 Constant Struct
struct nettle_cipher nettle_sha256 Constant Struct

These are all the hash functions that Nettle implements.

Node:Cipher functions, Next:, Previous:Hash functions, Up:Reference

Cipher functions

A cipher is a function that takes a message or plaintext and a secret key and transforms it to a ciphertext. Given only the ciphertext, but not the key, it should be hard to find the plaintext. Given matching pairs of plaintext and ciphertext, it should be hard to find the key.

There are two main classes of ciphers: Block ciphers and stream ciphers.

A block cipher can process data only in fixed size chunks, called blocks. Typical block sizes are 8 or 16 octets. To encrypt arbitrary messages, you usually have to pad it to an integral number of blocks, split it into blocks, and then process each block. The simplest way is to process one block at a time, independent of each other. That mode of operation is called ECB, Electronic Code Book mode. However, using ECB is usually a bad idea. For a start, plaintext blocks that are equal are transformed to ciphertext blocks that are equal; that leaks information about the plaintext. Usually you should apply the cipher is some feedback mode, CBC (Cipher Block Chaining) being one of the most popular. See Cipher Block Chaining, for information on how to apply CBC with Nettle.

A stream cipher can be used for messages of arbitrary length; a typical stream cipher is a keyed pseudo-random generator. To encrypt a plaintext message of n octets, you key the generator, generate n octets of pseudo-random data, and XOR it with the plaintext. To decrypt, regenerate the same stream using the key, XOR it to the ciphertext, and the plaintext is recovered.

Caution: The first rule for this kind of cipher is the same as for a One Time Pad: never ever use the same key twice.

A common misconception is that encryption, by itself, implies authentication. Say that you and a friend share a secret key, and you receive an encrypted message. You apply the key, and get a plaintext message that makes sense to you. Can you then be sure that it really was your friend that wrote the message you're reading? The answer is no. For example, if you were using a block cipher in ECB mode, an attacker may pick up the message on its way, and reorder, delete or repeat some of the blocks. Even if the attacker can't decrypt the message, he can change it so that you are not reading the same message as your friend wrote. If you are using a block cipher in CBC mode rather than ECB, or are using a stream cipher, the possibilities for this sort of attack are different, but the attacker can still make predictable changes to the message.

It is recommended to always use an authentication mechanism in addition to encrypting the messages. Popular choices are Message Authentication Codes like HMAC-SHA1 see Keyed hash functions, or digital signatures like RSA.

Some ciphers have so called "weak keys", keys that results in undesirable structure after the key setup processing, and should be avoided. In Nettle, the presence of weak keys for a cipher mean that the key setup function can fail, so you have to check its return value. In addition, the context struct has a field status, that is set to a non-zero value if key setup fails. When possible, avoid algorithm that have weak keys. There are several good ciphers that don't have any weak keys.

To encrypt a message, you first initialize a cipher context for encryption or decryption with a particular key. You then use the context to process plaintext or ciphertext messages. The initialization is known as called key setup. With Nettle, it is recommended to use each context struct for only one direction, even if some of the ciphers use a single key setup function that can be used for both encryption and decryption.

AES

AES is a quite new block cipher, specified by NIST as a replacement for the older DES standard. The standard is the result of a competition between cipher designers. The winning design, also known as RIJNDAEL, was constructed by Joan Daemen and Vincent Rijnmen.

Like all the AES candidates, the winning design uses a block size of 128 bits, or 16 octets, and variable key-size, 128, 192 and 256 bits (16, 24 and 32 octets) being the allowed key sizes. It does not have any weak keys. Nettle defines AES in <nettle/aes.h>.

struct aes_ctx Context struct

AES_BLOCK_SIZE Constant
The AES block-size, 16

AES_MIN_KEY_SIZE Constant

AES_MAX_KEY_SIZE Constant

AES_KEY_SIZE Constant
Default AES key size, 32

void aes_set_key (struct aes_ctx *ctx, unsigned length, const uint8_t *key) Function
Initialize the cipher. The same function is used for both encryption and decryption.

void aes_encrypt (struct aes_ctx *ctx, unsigned length, const uint8_t *dst, uint8_t *src) Function
Encryption function. length must be an integral multiple of the block size. If it is more than one block, the data is processed in ECB mode. src and dst may be equal, but they must not overlap in any other way.

void aes_decrypt (struct aes_ctx *ctx, unsigned length, const uint8_t *dst, uint8_t *src) Function
Analogous to aes_encrypt

ARCFOUR

ARCFOUR is a stream cipher, also known under the trade marked name RC4, and it is one of the fastest ciphers around. A problem is that the key setup of ARCFOUR is quite weak, you should never use keys with structure, keys that are ordinary passwords, or sequences of keys like "secret:1", "secret:2", ..... If you have keys that don't look like random bit strings, and you want to use ARCFOUR, always hash the key before feeding it to ARCFOUR. For example 

/* A more robust key setup function for ARCFOUR */
void
arcfour_set_key_hashed(struct arcfour_ctx *ctx,
                       unsigned length, const uint8_t *key)
{
  struct sha1_ctx hash;
  uint8_t digest[SHA1_DIGEST_SIZE];

  sha1_init(&hash);
  sha1_update(&hash, length, key);
  sha1_digest(&hash, SHA1_DIGEST_SIZE, digest);

  arcfour_set_key(ctx, SHA1_DIGEST_SIZE, digest);
}

Nettle defines ARCFOUR in <nettle/arcfour.h>.

struct arcfour_ctx Context struct

ARCFOUR_MIN_KEY_SIZE Constant
Minimum key size, 1

ARCFOUR_MAX_KEY_SIZE Constant
Maximum key size, 256

ARCFOUR_KEY_SIZE Constant
Default ARCFOUR key size, 16

void arcfour_set_key (struct arcfour_ctx *ctx, unsigned length, const uint8_t *key) Function
Initialize the cipher. The same function is used for both encryption and decryption.

void arcfour_crypt (struct arcfour_ctx *ctx, unsigned length, const uint8_t *key) Function
Encrypt some data. The same function is used for both encryption and decryption. Unlike the block ciphers, this function modifies the context, so you can split the data into arbitrary chunks and encrypt them one after another. The result is the same as if you had called arcfour_crypt only once with all the data.

CAST128

CAST-128 is a block cipher, specified in RFC 2144. It uses a 64 bit (8 octets) block size, and a variable key size of up to 128 bits. Nettle defines cast128 in <nettle/cast128.h>.

struct cast128_ctx Context struct

CAST128_BLOCK_SIZE Constant
The CAST128 block-size, 8

CAST128_MIN_KEY_SIZE Constant
Minimum CAST128 key size, 5

CAST128_MAX_KEY_SIZE Constant
Maximum CAST128 key size, 16

CAST128_KEY_SIZE Constant
Default CAST128 key size, 16

void cast128_set_key (struct cast128_ctx *ctx, unsigned length, const uint8_t *key) Function
Initialize the cipher. The same function is used for both encryption and decryption.

void cast128_encrypt (struct cast128_ctx *ctx, unsigned length, const uint8_t *dst, uint8_t *src) Function
Encryption function. length must be an integral multiple of the block size. If it is more than one block, the data is processed in ECB mode. src and dst may be equal, but they must not overlap in any other way.

void cast128_decrypt (struct cast128_ctx *ctx, unsigned length, const uint8_t *dst, uint8_t *src) Function
Analogous to cast128_encrypt

BLOWFISH

BLOWFISH is a block cipher designed by Bruce Schneier. It uses a block size of 64 bits (8 octets), and a variable key size, up to 448 bits. It has some weak keys. Nettle defines BLOWFISH in <nettle/blowfish.h>.

struct blowfish_ctx Context struct

BLOWFISH_BLOCK_SIZE Constant
The BLOWFISH block-size, 8

BLOWFISH_MIN_KEY_SIZE Constant
Minimum BLOWFISH key size, 8

BLOWFISH_MAX_KEY_SIZE Constant
Maximum BLOWFISH key size, 56

BLOWFISH_KEY_SIZE Constant
Default BLOWFISH key size, 16

int blowfish_set_key (struct blowfish_ctx *ctx, unsigned length, const uint8_t *key) Function
Initialize the cipher. The same function is used for both encryption and decryption. Returns 1 on success, and 0 if the key was weak. Calling blowfish_encrypt or blowfish_decrypt with a weak key will crash with an assert violation.

void blowfish_encrypt (struct blowfish_ctx *ctx, unsigned length, const uint8_t *dst, uint8_t *src) Function
Encryption function. length must be an integral multiple of the block size. If it is more than one block, the data is processed in ECB mode. src and dst may be equal, but they must not overlap in any other way.

void blowfish_decrypt (struct blowfish_ctx *ctx, unsigned length, const uint8_t *dst, uint8_t *src) Function
Analogous to blowfish_encrypt

DES

DES is the old Data Encryption Standard, specified by NIST. It uses a block size of 64 bits (8 octets), and a key size of 56 bits. However, the key bits are distributed over 8 octets, where the least significant bit of each octet is used for parity. A common way to use DES is to generate 8 random octets in some way, then set the least significant bit of each octet to get odd parity, and initialize DES with the resulting key.

The key size of DES is so small that keys can be found by brute force, using specialized hardware or lots of ordinary work stations in parallel. One shouldn't be using plain DES at all today, if one uses DES at all one should be using DES3 or "triple DES", see below.

DES also has some weak keys. Nettle defines DES in <nettle/des.h>.

struct des_ctx Context struct

DES_BLOCK_SIZE Constant
The DES block-size, 8

DES_KEY_SIZE Constant
DES key size, 8

int des_set_key (struct des_ctx *ctx, const uint8_t *key) Function
Initialize the cipher. The same function is used for both encryption and decryption. Returns 1 on success, and 0 if the key was weak or had bad parity. Calling des_encrypt or des_decrypt with a bad key will crash with an assert violation.

void des_encrypt (struct des_ctx *ctx, unsigned length, const uint8_t *dst, uint8_t *src) Function
Encryption function. length must be an integral multiple of the block size. If it is more than one block, the data is processed in ECB mode. src and dst may be equal, but they must not overlap in any other way.

void des_decrypt (struct des_ctx *ctx, unsigned length, const uint8_t *dst, uint8_t *src) Function
Analogous to des_encrypt

void des_fix_parity (unsigned length, uint8_t *dst, const uint8_t *src) Function
Adjusts the parity bits to match DES's requirements. You need this function if you have created a random-looking string by a key agreement protocol, and want to use it as a DES key. dst and src may be equal.

DES3

The inadequate key size of DES has already been mentioned. One way to increase the key size is to pipe together several DES boxes with independent keys. It turns out that using two DES ciphers is not as secure as one might think, even if the key size of the combination is a respectable 112 bits.

The standard way to increase DES's key size is to use three DES boxes. The mode of operation is a little peculiar: the middle DES box is wired in the reverse direction. To encrypt a block with DES3, you encrypt it using the first 56 bits of the key, then decrypt it using the middle 56 bits of the key, and finally encrypt it again using the last 56 bits of the key. This is known as "ede" triple-DES, for "encrypt-decrypt-encrypt".

The "ede" construction provides some backward compatibility, as you get plain single DES simply by feeding the same key to all three boxes. That should help keeping down the gate count, and the price, of hardware circuits implementing both plain DES and DES3.

DES3 has a key size of 168 bits, but just like plain DES, useless parity bits are inserted, so that keys are represented as 24 octets (192 bits). As a 112 bit key is large enough to make brute force attacks impractical, some applications uses a "two-key" variant of triple-DES. In this mode, the same key bits are used for the first and the last DES box in the pipe, while the middle box is keyed independently. The two-key variant is believed to be secure, i.e. there are no known attacks significantly better than brute force.

Naturally, it's simple to implement triple-DES on top of Nettle's DES functions. Nettle includes an implementation of three-key "ede" triple-DES, it is defined in the same place as plain DES, <nettle/des.h>.

struct des3_ctx Context struct

DES3_BLOCK_SIZE Constant
The DES3 block-size is the same as DES_BLOCK_SIZE, 8

DES3_KEY_SIZE Constant
DES key size, 24

int des3_set_key (struct des3_ctx *ctx, const uint8_t *key) Function
Initialize the cipher. The same function is used for both encryption and decryption. Returns 1 on success, and 0 if the key was weak or had bad parity. Calling des_encrypt or des_decrypt with a bad key will crash with an assert violation.

For random-looking strings, you can use des_fix_parity to adjust the parity bits before calling des3_set_key.

void des3_encrypt (struct des3_ctx *ctx, unsigned length, const uint8_t *dst, uint8_t *src) Function
Encryption function. length must be an integral multiple of the block size. If it is more than one block, the data is processed in ECB mode. src and dst may be equal, but they must not overlap in any other way.

void des3_decrypt (struct des3_ctx *ctx, unsigned length, const uint8_t *dst, uint8_t *src) Function
Analogous to des_encrypt

SERPENT

SERPENT is one of the AES finalists, designed by Ross Anderson, Eli Biham and Lars Knudsen. Thus, the interface and properties are similar to AES'. One peculiarity is that it is quite pointless to use it with anything but the maximum key size, smaller keys are just padded to larger ones. Nettle defines SERPENT in <nettle/serpent.h>.

struct serpent_ctx Context struct

SERPENT_BLOCK_SIZE Constant
The SERPENT block-size, 16

SERPENT_MIN_KEY_SIZE Constant
Minimum SERPENT key size, 16

SERPENT_MAX_KEY_SIZE Constant
Maximum SERPENT key size, 32

SERPENT_KEY_SIZE Constant
Default SERPENT key size, 32

void serpent_set_key (struct serpent_ctx *ctx, unsigned length, const uint8_t *key) Function
Initialize the cipher. The same function is used for both encryption and decryption.

void serpent_encrypt (struct serpent_ctx *ctx, unsigned length, const uint8_t *dst, uint8_t *src) Function
Encryption function. length must be an integral multiple of the block size. If it is more than one block, the data is processed in ECB mode. src and dst may be equal, but they must not overlap in any other way.

void serpent_decrypt (struct serpent_ctx *ctx, unsigned length, const uint8_t *dst, uint8_t *src) Function
Analogous to serpent_encrypt

TWOFISH

Another AES finalist, this one designed by Bruce Schneier and others. Nettle defines it in <nettle/twofish.h>.

struct twofish_ctx Context struct

TWOFISH_BLOCK_SIZE Constant
The TWOFISH block-size, 16

TWOFISH_MIN_KEY_SIZE Constant
Minimum TWOFISH key size, 16

TWOFISH_MAX_KEY_SIZE Constant
Maximum TWOFISH key size, 32

TWOFISH_KEY_SIZE Constant
Default TWOFISH key size, 32

void twofish_set_key (struct twofish_ctx *ctx, unsigned length, const uint8_t *key) Function
Initialize the cipher. The same function is used for both encryption and decryption.

void twofish_encrypt (struct twofish_ctx *ctx, unsigned length, const uint8_t *dst, uint8_t *src) Function
Encryption function. length must be an integral multiple of the block size. If it is more than one block, the data is processed in ECB mode. src and dst may be equal, but they must not overlap in any other way.

void twofish_decrypt (struct twofish_ctx *ctx, unsigned length, const uint8_t *dst, uint8_t *src) Function
Analogous to twofish_encrypt

struct nettle_cipher

Nettle includes a struct including information about some of the more regular cipher functions. It should be considered a little experimental, but can be useful for applications that need a simple way to handle various algorithms. Nettle defines these structs in <nettle/nettle-meta.h>.

struct nettle_cipher name context_size block_size key_size set_encrypt_key set_decrypt_key encrypt decrypt Meta struct
The last four attributes are function pointers, of types nettle_set_key_func and nettle_crypt_func. The first argument to these functions is a void * pointer to a context struct, which is of size context_size.

struct nettle_cipher nettle_aes128 Constant Struct
struct nettle_cipher nettle_aes192 Constant Struct
struct nettle_cipher nettle_aes256 Constant Struct
struct nettle_cipher nettle_arcfour128 Constant Struct
struct nettle_cipher nettle_cast128 Constant Struct
struct nettle_cipher nettle_serpent128 Constant Struct
struct nettle_cipher nettle_serpent192 Constant Struct
struct nettle_cipher nettle_serpent256 Constant Struct
struct nettle_cipher nettle_twofish128 Constant Struct
struct nettle_cipher nettle_twofish192 Constant Struct
struct nettle_cipher nettle_twofish256 Constant Struct

Nettle includes such structs for all the regular ciphers, i.e. ones without weak keys or other oddity.

Node:Cipher Block Chaining, Next:, Previous:Cipher functions, Up:Reference

Cipher Block Chaining

When using CBC mode, plaintext blocks are not encrypted independently of each other, like in Electronic Cook Book mode. Instead, when encrypting a block in CBC mode, the previous ciphertext block is XOR:ed with the plaintext before it is fed to the block cipher. When encrypting the first block, a random block called an IV, or Initialization Vector, is used as the "previous ciphertext block". The IV should be chosen randomly, but it need not be kept secret, and can even be transmitted in the clear together with the encrypted data.

In symbols, if E_k is the encryption function of a block cipher, and IV is the initialization vector, then n plaintext blocks M_1,... M_n are transformed into n ciphertext blocks C_1,... C_n as follows: 

C_1 = E_k(IV  XOR M_1)
C_2 = E_k(C_1 XOR M_2)

...

C_n = E_k(C_(n-1) XOR M_n)

Nettle includes a few utility functions for applying a block cipher in Cipher Block Chaining (CBC) mode. The functions uses void * to pass cipher contexts around.

void cbc_encrypt (void *ctx, void (*f)(), unsigned block_size, uint8_t *iv, unsigned length, uint8_t *dst, const uint8_t *src) Function
void cbc_decrypt (void *ctx, void (*f)(), unsigned block_size, uint8_t *iv, unsigned length, uint8_t *dst, const uint8_t *src) Function

Applies the encryption or decryption function f in CBC mode. The function f is really typed as

void f (void *ctx, unsigned length, uint8_t dst, const uint8_t *src),

and the cbc_encrypt and cbc_decrypt functions pass their argument ctx on to f.

There are also some macros to help use these functions correctly.

CBC_CTX (context_type, block_size) Macro
Expands into 
{
   context_type ctx;
   uint8_t iv[block_size];
}

It can be used to define a CBC context struct, either directly, 

struct CBC_CTX(struct aes_ctx, AES_BLOCK_SIZE) ctx;

or to give it a struct tag, 

struct aes_cbc_ctx CBC_CTX (struct aes_ctx, AES_BLOCK_SIZE);

CBC_SET_IV (ctx, iv) Macro
First argument is a pointer to a context struct as defined by CBC_CTX, and the second is a pointer to an Initialization Vector (IV) that is copied into that context.

CBC_ENCRYPT (ctx, f, length, dst, src) Macro
CBC_DECRYPT (ctx, f, length, dst, src) Macro
A simpler way to invoke cbc_encrypt and cbc_decrypt. The first argument is a pointer to a context struct as defined by CBC_CTX, and the second argument is an encryption or decryption function following Nettle's conventions. The last three arguments define the source and destination area for the operation.

These macros use some tricks to make the compiler display a warning if the types of f and ctx don't match, e.g. if you try to use an struct aes_ctx context with the des_encrypt function.

Node:Keyed hash functions, Next:, Previous:Cipher Block Chaining, Up:Reference

Keyed Hash Functions

A keyed hash function, or Message Authentication Code (MAC) is a function that takes a key and a message, and produces fixed size MAC. It should be hard to compute a message and a matching MAC without knowledge of the key. It should also be hard to compute the key given only messages and corresponding MACs.

Keyed hash functions are useful primarily for message authentication, when the Alice and Bob shares a secret: The sender, Alice, computes the MAC and attaches it to the message. The receiver, Bob, also computes the MAC of the message, using the same key, and compares that to Alice's value. If they match, Bob can be assured that the message has not been modified on it's way from Alice.

However, unlike digital signatures, this assurance is not transferable. Bob can't show the message and the MAC to a third party and prove that Alice sent that message. Not even if he gives away the key to the third party. The reason is that the same key is used on both sides, and anyone knowing the key can create a correct MAC for any message. If Bob believes that only he and Alice knows the key, and he knows that he didn't attach a MAC to a particular message, he knows it must be Alice who did it. However, the third party can't distinguish between MAC created by Alice and one created by Bob.

Keyed hash functions are typically a lot faster than digital signatures as well.

HMAC

One can build keyed hash functions from ordinary hash functions. Older constructions simply concatenate secret key and message and hashes that, but such constructions have weaknesses. A better construction is HMAC, described in RFC 2104.

For an underlying hash function H, with digest size l and internal block size b, HMAC-H is constructed as follows: From a given key k, two distinct subkeys k_i and k_o are constructed, both of length b. The HMAC-H of a message m is then computed as H(k_o | H(k_i | m)), where | denotes string concatenation.

HMAC keys can be of any length, but it is recommended to use keys of length l, the digest size of the underlying hash function H. Keys that are longer than b are shortened to length l by hashing with H, so arbitrarily long keys aren't very useful.

Nettle's HMAC functions are defined in <nettle/hmac.h>. There are abstract functions that use a pointer to a struct nettle_hash to represent the underlying hash function and void * pointers that point to three different context structs for that hash function. There are also concrete functions for HMAC-MD5, HMAC-SHA1, and HMAC-SHA256. First, the abstract functions:

void hmac_set_key (void *outer, void *inner, void *state, const struct nettle_hash *H, unsigned length, const uint8_t *key) Function
Initializes the three context structs from the key. The outer and inner contexts corresponds to the subkeys k_o and k_i. state is used for hashing the message, and is initialized as a copy of the inner context.

void hmac_update (void *state, const struct nettle_hash *H, unsigned length, const uint8_t *data) Function
This function is called zero or more times to process the message. Actually, hmac_update(state, H, length, data) is equivalent to H->update(state, length, data), so if you wish you can use the ordinary update function of the underlying hash function instead.

void hmac_digest (const void *outer, const void *inner, void *state, const struct nettle_hash *H, unsigned length, uint8_t *digest) Function
Extracts the MAC of the message, writing it to digest. outer and inner are not modified. length is usually equal to H->digest_size, but if you provide a smaller value, only the first length octets of the MAC are written.

This function also resets the state context so that you can start over processing a new message (with the same key).

Like for CBC, there are some macros to help use these functions correctly.

HMAC_CTX (type) Macro
Expands into 
{
   type outer;
   type inner;
   type state;
}

It can be used to define a HMAC context struct, either directly, 

struct HMAC_CTX(struct md5_ctx) ctx;

or to give it a struct tag, 

struct hmac_md5_ctx HMAC_CTX (struct md5_ctx);

HMAC_SET_KEY (ctx, H, length, key) Macro
ctx is a pointer to a context struct as defined by HMAC_CTX, H is a pointer to a const struct nettle_hash describing the underlying hash function (so it must match the type of the components of ctx). The last two arguments specify the secret key.

HMAC_DIGEST (ctx, H, length, digest) Macro
ctx is a pointer to a context struct as defined by HMAC_CTX, H is a pointer to a const struct nettle_hash describing the underlying hash function. The last two arguments specify where the digest is written.

Note that there is no HMAC_UPDATE macro; simply call hmac_update function directly, or the update function of the underlying hash function.

Concrete HMAC functions

Now we come to the specialized HMAC functions, which are easier to use than the general HMAC functions.

HMAC-MD5

struct hmac_md5_ctx Context struct

void hmac_md5_set_key (struct hmac_md5_ctx *ctx, unsigned key_length, const uint8_t *key) Function
Initializes the context with the key.

void hmac_md5_update (struct hmac_md5_ctx *ctx, unsigned length, const uint8_t *data) Function
Process some more data.

void hmac_md5_digest (struct hmac_md5_ctx *ctx, unsigned length, uint8_t *digest) Function
Extracts the MAC, writing it to digest. length may be smaller than MD5_DIGEST_SIZE, in which case only the first length octets of the MAC are written.

This function also resets the context for processing new messages, with the same key.

HMAC-SHA1

struct hmac_sha1_ctx Context struct

void hmac_sha1_set_key (struct hmac_sha1_ctx *ctx, unsigned key_length, const uint8_t *key) Function
Initializes the context with the key.

void hmac_sha1_update (struct hmac_sha1_ctx *ctx, unsigned length, const uint8_t *data) Function
Process some more data.

void hmac_sha1_digest (struct hmac_sha1_ctx *ctx, unsigned length, uint8_t *digest) Function
Extracts the MAC, writing it to digest. length may be smaller than SHA1_DIGEST_SIZE, in which case only the first length octets of the MAC are written.

This function also resets the context for processing new messages, with the same key.

HMAC-SHA256

struct hmac_sha256_ctx Context struct

void hmac_sha256_set_key (struct hmac_sha256_ctx *ctx, unsigned key_length, const uint8_t *key) Function
Initializes the context with the key.

void hmac_sha256_update (struct hmac_sha256_ctx *ctx, unsigned length, const uint8_t *data) Function
Process some more data.

void hmac_sha256_digest (struct hmac_sha256_ctx *ctx, unsigned length, uint8_t *digest) Function
Extracts the MAC, writing it to digest. length may be smaller than SHA256_DIGEST_SIZE, in which case only the first length octets of the MAC are written.

This function also resets the context for processing new messages, with the same key.

Node:Public-key algorithms, Next:, Previous:Keyed hash functions, Up:Reference

Public-key algorithms

Nettle uses GMP, the GNU bignum library, for all calculations with large numbers. In order to use the public-key features of Nettle, you must install GMP, at least version 3.0, before compiling Nettle, and you need to link your programs with -lgmp.

The concept of Public-key encryption and digital signatures was discovered by Whitfield Diffie and Martin E. Hellman and described in a paper 1976. In traditional, "symmetric", cryptography, sender and receiver share the same keys, and these keys must be distributed in a secure way. And if there are many users or entities that need to communicate, each pair needs a shared secret key known by nobody else.

Public-key cryptography uses trapdoor one-way functions. A one-way function is a function F such that it is easy to compute the value F(x) for any x, but given a value y, it is hard to compute a corresponding x such that y = F(x). Two examples are cryptographic hash functions, and exponentiation in certain groups.

A trapdoor one-way function is a function F that is one-way, unless one knows some secret information about F. If one knows the secret, it is easy to compute both F and it's inverse. If this sounds strange, look at the RSA example below.

Two important uses for one-way functions with trapdoors are public-key encryption, and digital signatures. Of these, I won't say more about public-key encryption, as that isn't yet supported by Nettle. So the rest of this chapter is about digital signatures.

To use a digital signature algorithm, one must first create a key-pair: A public key and a corresponding private key. The private key is used to sign messages, while the public key is used for verifying that that signatures and messages match. Some care must be taken when distributing the public key; it need not be kept secret, but if a bad guy is able to replace it (in transit, or in some user's list of known public keys), bad things may happen.

There are two operations one can do with the keys. The signature operation takes a message and a private key, and creates a signature for the message. A signature is some string of bits, usually at most a few thousand bits or a few hundred octets. Unlike paper-and-ink signatures, the digital signature depends on the message, so one can't cut it out of context and glue it to a different message.

The verification operation takes a public key, a message, and a string that is claimed to be a signature on the message, and returns true or false. If it returns true, that means that the three input values matched, and the verifier can be sure that someone went through with the signature operation on that very message, and that the "someone" also knows the private key corresponding to the public key.

The desired properties of a digital signature algorithm are as follows: Given the public key and pairs of messages and valid signatures on them, it should be hard to compute the private key, and it should also be hard to create a new message and signature that is accepted by the verification operation.

Besides signing meaningful messages, digital signatures can be used for authorization. A server can be configured with a public key, such that any client that connects to the service is given a random nonce message. If the server gets a reply with a correct signature matching the nonce message and the configured public key, the client is granted access. So the configuration of the server can be understood as "grant access to whoever knows the private key corresponding to this particular public key, and to no others".

RSA

The RSA was the first practical digital signature algorithm that was constructed. It was described 1978 in a paper by Ronald Rivest, Adi Shamir and L.M. Adleman, and the technique was also patented in 1983. The patent expired on September 20, 2000, and since that day, RSA can be used freely.

It's remarkably simple to describe the trapdoor function behind RSA. The "one-way"-function used is 

F(x) = x^e mod n

I.e. raise x to the e:th power, while discarding all multiples of n. The pair of numbers n and e is the public key. e can be quite small, even e = 3 has been used, although slightly larger numbers are recommended. n should be about 1000 bits or larger.

If n is large enough, and properly chosen, the inverse of F, the computation of e:th roots modulo n, is very difficult. But, where's the trapdoor?

Let's first look at how RSA key-pairs are generated. First n is chosen as the product of two large prime numbers p and q of roughly the same size (so if n is 1000 bits, p and q are about 500 bits each). One also computes the number phi = (p-1)(q-1), in mathematical speak, phi is the order of the multiplicative group of integers modulo n.

Next, e is chosen. It must have no factors in common with phi (in particular, it must be odd), but can otherwise be chosen more or less randomly. e = 65537 is a popular choice, because it makes raising to the e:th power particularly efficient, and being prime, it usually has no factors common with phi.

Finally, a number d, d < n is computed such that e d mod phi = 1. It can be shown that such a number exists (this is why e and phi must have no common factors), and that for all x, 

(x^e)^d mod n = x^(ed) mod n = (x^d)^e mod n = x

Using Euclid's algorithm, d can be computed quite easily from phi and e. But it is still hard to get d without knowing phi, which depends on the factorization of n.

So d is the trapdoor, if we know d and y = F(x), we can recover x as y^d mod n. d is also the private half of the RSA key-pair.

The most common signature operation for RSA is defined in PKCS#1, a specification by RSA Laboratories. The message to be signed is first hashed using a cryptographic hash function, e.g. MD5 or SHA1. Next, some padding, the ASN.1 "Algorithm Identifier" for the hash function, and the message digest itself, are concatenated and converted to a number x. The signature is computed from x and the private key as s = x^d mod n1. The signature, s is a number of about the same size of n, and it usually encoded as a sequence of octets, most significant octet first.

The verification operation is straight-forward, x is computed from the message in the same way as above. Then s^e mod n is computed, the operation returns true if and only if the result equals x.

Nettle's RSA support

Nettle represents RSA keys using two structures that contain large numbers (of type mpz_t).

rsa_public_key size n e Context struct
size is the size, in octets, of the modulo, and is used internally. n and e is the public key.

rsa_private_key size d p q a b c Context struct
size is the size, in octets, of the modulo, and is used internally. d is the secret exponent, but it is not actually used when signing. Instead, the factors p and q, and the parameters a, b and c are used. They are computed from p, q and d such that a e mod (p - 1) = 1, b e mod (q - 1) = 1, c q mod p= 1.

Before use, these structs must be initialized by calling one of

void rsa_init_public_key (struct rsa_public_key *pub) Function
void rsa_init_private_key (struct rsa_private_key *key) Function
Calls mpz_init on all numbers in the key struct.

and when finished with them, the space for the numbers must be deallocated by calling one of

void rsa_clear_public_key (struct rsa_public_key *pub) Function
void rsa_clear_private_key (struct rsa_private_key *key) Function
Calls mpz_clear on all numbers in the key struct.

In general, Nettle's rsa functions deviates from Nettle's "no memory allocation"-policy. Space for all the numbers, both in the key structs above, and temporaries, are allocated dynamically. For information on how to customize allocation, see See GMP Allocation.

When you have assigned values to the attributes of a key, you must call

int rsa_prepare_public_key (struct rsa_public_key *pub) Function
int rsa_prepare_private_key (struct rsa_private_key *key) Function
Computes the octet size of the key (stored in the size attribute, and may also do other basic sanity checks. Returns one if successful, or zero if the key can't be used, for instance if the modulo is smaller than the minimum size specified by PKCS#1.

Before signing or verifying a message, you first hash it with the appropriate hash function. You pass the hash function's context struct to the rsa function, and it will extract the message digest and do the rest of the work.

Creation and verification of signatures is done with the following functions:

void rsa_md5_sign (struct rsa_private_key *key, struct md5_ctx *hash, mpz_t signature) Function
void rsa_sha1_sign (struct rsa_private_key *key, struct sha1_ctx *hash, mpz_t signature) Function
The signature is stored in signature (which must have been mpz_init:ed earlier). The hash context is reset so that it can be used for new messages.

int rsa_md5_verify (struct rsa_public_key *key, struct md5_ctx *hash, const mpz_t signature) Function
int rsa_sha1_verify (struct rsa_public_key *key, struct sha1_ctx *hash, const mpz_t signature) Function
Returns 1 if the signature is valid, or 0 if it isn't. In either case, the hash context is reset so that it can be used for new messages.

If you need to use the RSA trapdoor, the private key, in a way that isn't supported by the above functions Nettle also includes a function that computes x^d mod n and nothing more, using the CRT optimization.

void rsa_compute_root (struct rsa_private_key *key, mpz_t x, const mpz_t m) Function
Computes x = m^d, efficiently.

At last, how do you create new keys?

int rsa_generate_keypair (struct rsa_public_key *pub, struct rsa_private_key *key, void *random_ctx, nettle_random_func random, void *progress_ctx, nettle_progress_func progress, unsigned n_size, unsigned e_size); Function
There are lots of parameters. pub and key is where the resulting key pair is stored. The structs should be initialized, but you don't need to call rsa_prepare_public_key or rsa_prepare_private_key after key generation.

random_ctx and random is a randomness generator. random(random_ctx, length, dst) should generate length random octets and store them at dst. For advice, see See Randomness.

progress and progress_ctx can be used to get callbacks during the key generation process, in order to uphold an illusion of progress. progress can be NULL, in that case there are no callbacks.

size_n is the desired size of the modulo, in bits. If size_e is non-zero, it is the desired size of the public exponent and a random exponent of that size is selected. But if e_size is zero, it is assumed that the caller has already chosen a value for e, and stored it in pub.

Returns 1 on success, and 0 on failure. The function can fail for example if if n_size is too small, or if e_size is zero and pub->e is an even number.

Node:Randomness, Next:, Previous:Public-key algorithms, Up:Reference

Randomness

A crucial ingredient in many cryptographic contexts is randomness: Let p be a random prime, choose a random initialization vector iv, a random key k and a random exponent e, etc. In the theories, it is assumed that you have plenty of randomness around. If this assumption is not true in practice, systems that are otherwise perfectly secure, can be broken. Randomness has often turned out to be the weakest link in the chain.

In non-cryptographic applications, such as games as well as scientific simulation, a good randomness generator usually means a generator that has good statistical properties, and is seeded by some simple function of things like the current time, process id, and host name.

However, such a generator is inadequate for cryptography, for at least two reasons:

A randomness generator that is used for cryptographic purposes must have better properties. Let's first look at the seeding, as the issues here are mostly independent of the rest of the generator. The initial state of the generator (its seed) must be unguessable by the attacker. So what's unguessable? It depends on what the attacker already knows. The concept used in information theory to reason about such things is called "entropy", or "conditional entropy" (not to be confused with the thermodynamic concept with the same name). A reasonable requirement is that the seed contains a conditional entropy of at least some 80-100 bits. This property can be explained as follows: Allow the attacker to ask n yes-no-questions, of his own choice, about the seed. If the attacker, using this question-and-answer session, as well as any other information he knows about the seeding process, still can't guess the seed correctly, then the conditional entropy is more than n bits.

Let's look at an example. Say information about timing of received network packets is used in the seeding process. If there is some random network traffic going on, this will contribute some bits of entropy or "unguessability" to the seed. However, if the attacker can listen in to the local network, or if all but a small number of the packets were transmitted by machines that the attacker can monitor, this additional information makes the seed easier for the attacker to figure out. Even if the information is exactly the same, the conditional entropy, or unguessability, is smaller for an attacker that knows some of it already before the hypothetical question-and-answer session.

Seeding of good generators is usually based on several sources. The key point here is that the amount of unguessability that each source contributes, depends on who the attacker is. Some sources that have been used are:

High resolution timing of i/o activities
Such as completed blocks from spinning hard disks, network packets, etc. Getting access to such information is quite system dependent, and not all systems include suitable hardware. If available, it's one of the better randomness source one can find in a digital, mostly predictable, computer.
User activity
Timing and contents of user interaction events is another popular source that is available for interactive programs (even if I suspect that it is sometimes used in order to make the user feel good, not because the quality of the input is needed or used properly). Obviously, not available when a machine is unattended. Also beware of networks: User interaction that happens across a long serial cable, TELNET session, or even SSH session may be visible to an attacker, in full or partially.
Audio input
Any room, or even a microphone input that's left unconnected, is a source of some random background noise, which can be fed into the seeding process.
Specialized hardware
Hardware devices with the sole purpose of generating random data have been designed. They range from radioactive samples with an attached Geiger counter, to amplification of the inherent noise in electronic components such as diodes and resistors, to low-frequency sampling of chaotic systems. Hashing successive images of a Lava lamp is a spectacular example of the latter type.
Secret information
Secret information, such as user passwords or keys, or private files stored on disk, can provide some unguessability. A problem is that if the information is revealed at a later time, the unguessability vanishes. Another problem is that this kind of information tends to be fairly constant, so if you rely on it and seed your generator regularly, you risk constructing almost similar seeds or even constructing the same seed more than once.

For all practical sources, it's difficult but important to provide a reliable lower bound on the amount of unguessability that it provides. Two important points are to make sure that the attacker can't observe your sources (so if you like the Lava lamp idea, remember that you have to get your own lamp, and not put it by a window or anywhere else where strangers can see it), and that hardware failures are detected. What if the bulb in the Lava lamp, which you keep locked into a cupboard following the above advice, breaks after a few months?

So let's assume that we have been able to find an unguessable seed, which contains at least 80 bits of conditional entropy, relative to all attackers that we care about (typically, we must at the very least assume that no attacker has root privileges on our machine).

How do we generate output from this seed, and how much can we get? Some generators (notably the Linux /dev/random generator) tries to estimate available entropy and restrict the amount of output. The goal is that if you read 128 bits from /dev/random, you should get 128 "truly random" bits. This is a property that is useful in some specialized circumstances, for instance when generating key material for a one time pad, or when working with unconditional blinding, but in most cases, it doesn't matter much. For most application, there's no limit on the amount of useful "random" data that we can generate from a small seed; what matters is that the seed is unguessable and that the generator has good cryptographic properties.

At the heart of all generators lies its internal state. Future output is determined by the internal state alone. Let's call it the generator's key. The key is initialized from the unguessable seed. Important properties of a generator are:


Key-hiding
An attacker observing the output should not be able to recover the generator's key.
Independence of outputs
Observing some of the output should not help the attacker to guess previous or future output.
Forward secrecy
Even if an attacker compromises the generator's key, he should not be able to guess the generator output before the key compromise.
Recovery from key compromise
If an attacker compromises the generator's key, he can compute all future output. This is inevitable if the generator is seeded only once, at startup. However, the generator can provide a reseeding mechanism, to achieve recovery from key compromise. More precisely: If the attacker compromises the key at a particular time t_1, there is another later time t_2, such that if the attacker observes all output generated between t_1 and t_2, he still can't guess what output is generated after t_2.

Nettle includes one randomness generator that is believed to have all the above properties, and two simpler ones.

ARCFOUR, like any stream cipher, can be used as a randomness generator. Its output should be of reasonable quality, if the seed is hashed properly before it is used with arcfour_set_key. There's no single natural way to reseed it, but if you need reseeding, you should be using Yarrow instead.

The "lagged Fibonacci" generator in <nettle/knuth-lfib.h> is a fast generator with good statistical properties, but is not for cryptographic use, and therefore not documented here. It is included mostly because the Nettle test suite needs to generate some test data from a small seed.

The recommended generator to use is Yarrow, described below.

Yarrow

Yarrow is a family of pseudo-randomness generators, designed for cryptographic use, by John Kelsey, Bruce Schneier and Niels Ferguson. Yarrow-160 is described in a paper at <http://www.counterpane.com/yarrow.html>, and it uses SHA1 and triple-DES, and has a 160-bit internal state. Nettle implements Yarrow-256, which is similar, but uses SHA256 and AES to get an internal state of 256 bits.

Yarrow was an almost finished project, the paper mentioned above is the closest thing to a specification for it, but some smaller details are left out. There is no official reference implementation or test cases. This section includes an overview of Yarrow, but for the details of Yarrow-256, as implemented by Nettle, you have to consult the source code. Maybe a complete specification can be written later.

Yarrow can use many sources (at least two are needed for proper reseeding), and two randomness "pools", referred to as the "slow pool" and the "fast pool". Input from the sources is fed alternatingly into the two pools. When one of the sources has contributed 100 bits of entropy to the fast pool, a "fast reseed" happens and the fast pool is mixed into the internal state. When at least two of the sources have contributed at least 160 bits each to the slow pool, a "slow reseed" takes place. The contents of both pools are mixed into the internal state. These procedures should ensure that the generator will eventually recover after a key compromise.

The output is generated by using AES to encrypt a counter, using the generator's current key. After each request for output, another 256 bits are generated which replace the key. This ensures forward secrecy.

Yarrow can also use a seed file to save state across restarts. Yarrow is seeded by either feeding it the contents of the previous seed file, or feeding it input from its sources until a slow reseed happens.

Nettle defines Yarrow-256 in <nettle/yarrow.h>.

struct yarrow256_ctx Context struct

struct yarrow_source Context struct
Information about a single source.

YARROW256_SEED_FILE_SIZE Constant
The size of the Yarrow-256 seed file.

void yarrow256_init (struct yarrow256_ctx *ctx, unsigned nsources, struct yarrow_source *sources) Function
Initializes the yarrow context, and its nsources sources. It's possible to use call it with nsources=0 and sources=NULL, if you don't need the update features.

void yarrow256_seed (struct yarrow256_ctx *ctx, unsigned length, uint8_t *seed_file) Function
Seeds Yarrow-256 from a previous seed file. length should be at least YARROW256_SEED_FILE_SIZE, but it can be larger.

The generator will trust you that the seed_file data really is unguessable. After calling this function, you must overwrite the old seed file with the contents of ctx->seed_file. If it's possible for several processes to read the seed file at about the same time, access must be coordinated, for example using lock files.

int yarrow256_update (struct yarrow256_ctx *ctx, unsigned source, unsigned entropy, unsigned length, const uint8_t *data) Function
Updates the generator with data from source SOURCE (an index that must be smaller than the number of sources). entropy is your estimated lower bound for the entropy in the data, measured in bits. Calling update with zero entropy is always safe, no matter if the data is random or not.

Returns 1 if a reseed happened, in which case the seed file can be overwritten with the contents of ctx->seed_file. Otherwise, the function returns 0.

void yarrow256_random (struct yarrow256_ctx *ctx, unsigned length, uint8_t *dst) Function
Generates length octets of output. The generator must be seeded before you call this function.

If you don't need forward secrecy, e.g. if you need non-secret randomness for initialization vectors or padding, you can gain some efficiency by buffering, calling this function for reasonably large blocks of data, say 100-1000 octets at a time.

int yarrow256_is_seeded (struct yarrow256_ctx *ctx) Function
Returns 1 if the generator is seeded and ready to generate output, otherwise 0.

unsigned yarrow256_needed_sources (struct yarrow256_ctx *ctx) Function
Returns the number of sources that must reach the threshold before a slow reseed will happen. Useful primarily when the generator is unseeded.

void yarrow256_force_reseed (struct yarrow256_ctx *ctx) Function
Causes a slow reseed to take place immediately, regardless of the current entropy estimates of the two pools. Use with care.

Nettle includes an entropy estimator for one kind of input source: User keyboard input.

struct yarrow_key_event_ctx Context struct
Information about recent key events.

void yarrow_key_event_init (struct yarrow_key_event_ctx *ctx) Function
Initializes the context.

unsigned yarrow_key_event_estimate (struct yarrow_key_event_ctx *ctx, unsigned key, unsigned time) Function
key is the id of the key (ASCII value, hardware key code, X keysym, ... it doesn't matter), and time is the timestamp of the event. The time must be given in units matching the resolution by which you read the clock. If you read the clock with microsecond precision, time should be provided in units of microseconds. But if you use gettimeofday on a typical Unix system where the clock ticks 10 or so microseconds at a time, time should be given in units of 10 microseconds.

Returns an entropy estimate, in bits, suitable for calling yarrow256_update. Usually, 0, 1 or 2 bits.

Node:Miscellaneous functions, Next:, Previous:Randomness, Up:Reference

Miscellaneous functions

uint8_t * memxor (uint8_t *dst, const uint8_t *src, size_t n) Function
XOR:s the source area on top of the destination area. The interface doesn't follow the Nettle conventions, because it is intended to be similar to the ANSI-C memcpy function.

memxor is declared in <nettle/memxor.h>.

Node:Compatibility functions, Previous:Miscellaneous functions, Up:Reference

Compatibility functions

For convenience, Nettle includes alternative interfaces to some algorithms, for compatibility with some other popular crypto toolkits. These are not fully documented here; refer to the source or to the documentation for the original implementation.

MD5 is defined in [RFC 1321], which includes a reference implementation. Nettle defines a compatible interface to MD5 in <nettle/md5-compat.h>. This file defines the typedef MD5_CTX, and declares the functions MD5Init, MD5Update and MD5Final.

Eric Young's "libdes" (also part of OpenSSL) is a quite popular DES implementation. Nettle includes a subset if it's interface in <nettle/des-compat.h>. This file defines the typedefs des_key_schedule and des_cblock, two constants DES_ENCRYPT and DES_DECRYPT, and declares one global variable des_check_key, and the functions des_cbc_cksum des_cbc_encrypt, des_ecb2_encrypt, des_ecb3_encrypt, des_ecb_encrypt, des_ede2_cbc_encrypt, des_ede3_cbc_encrypt, des_is_weak_key, des_key_sched, des_ncbc_encrypt des_set_key, and des_set_odd_parity.

Node:Nettle soup, Next:, Previous:Reference, Up:Top

Traditional Nettle Soup

For the serious nettle hacker, here is a recipe for nettle soup. 4 servings

Gather 1 liter fresh nettles. Use gloves! Small, tender shoots are preferable but the tops of larger nettles can also be used.

Rinse the nettles very well. Boil them for 10 minutes in lightly salted water. Strain the nettles and save the water. Hack the nettles. Melt the butter and mix in the flour. Dilute with bouillon and the nettle-water you save earlier. Add the hacked nettles. If you wish you can add some milk or cream at this stage. Bring to a boil and let boil for a few minutes. Season with salt and pepper.

Serve with boiled egg-halves.

Node:Installation, Next:, Previous:Nettle soup, Up:Top

Installation

Nettle uses autoconf and automake. To build it, unpack the source and run 

./configure
make
make check
make install

to install in the default location, /usr/local. The library is installed in /use/local/lib/libnettle.a and the include files are installed in /use/local/include/nettle/.

Only static libraries are installed.

Node:Index, Previous:Installation, Up:Top

Function and Concept Index

Footnotes

  1. Actually, the computation is not done like this, it is done more efficiently using p, q and the Chinese remainder theorem (CRT). But the result is the same.