tiny_keccak/parallel_hash.rs
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206
use crate::{left_encode, right_encode, CShake, Hasher, IntoXof, Xof};
#[derive(Clone)]
struct UnfinishedState {
state: CShake,
absorbed: usize,
}
struct Suboutout {
state: [u8; 64],
size: usize,
}
impl Suboutout {
fn security(bits: usize) -> Suboutout {
Suboutout {
state: [0u8; 64],
// 128 => 32, 256 => 64
size: bits / 4,
}
}
#[inline]
fn as_bytes(&self) -> &[u8] {
&self.state[..self.size]
}
#[inline]
fn as_bytes_mut(&mut self) -> &mut [u8] {
&mut self.state[..self.size]
}
}
/// The `ParallelHash` hash functions defined in [`SP800-185`].
///
/// The purpose of `ParallelHash` is to support the efficient hashing of very long strings, by
/// taking advantage of the parallelism available in modern processors. `ParallelHash` supports the
/// [`128-bit`] and [`256-bit`] security strengths, and also provides variable-length output.
///
/// # Usage
///
/// ```toml
/// [dependencies]
/// tiny-keccak = { version = "2.0.0", features = ["parallel_hash"] }
/// ```
///
/// [`SP800-185`]: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-185.pdf
/// [`128-bit`]: struct.ParallelHash.html#method.v128
/// [`256-bit`]: struct.ParallelHash.html#method.v256
#[derive(Clone)]
pub struct ParallelHash {
state: CShake,
block_size: usize,
bits: usize,
blocks: usize,
unfinished: Option<UnfinishedState>,
}
impl ParallelHash {
/// Creates new [`ParallelHash`] hasher with a security level of 128 bits.
///
/// [`ParallelHash`]: struct.ParallelHash.html
pub fn v128(custom_string: &[u8], block_size: usize) -> ParallelHash {
ParallelHash::new(custom_string, block_size, 128)
}
/// Creates new [`ParallelHash`] hasher with a security level of 256 bits.
///
/// [`ParallelHash`]: struct.ParallelHash.html
pub fn v256(custom_string: &[u8], block_size: usize) -> ParallelHash {
ParallelHash::new(custom_string, block_size, 256)
}
fn new(custom_string: &[u8], block_size: usize, bits: usize) -> ParallelHash {
let mut state = CShake::new(b"ParallelHash", custom_string, bits);
state.update(left_encode(block_size).value());
ParallelHash {
state,
block_size,
bits,
blocks: 0,
unfinished: None,
}
}
}
impl Hasher for ParallelHash {
fn update(&mut self, mut input: &[u8]) {
if let Some(mut unfinished) = self.unfinished.take() {
let to_absorb = self.block_size - unfinished.absorbed;
if input.len() >= to_absorb {
unfinished.state.update(&input[..to_absorb]);
input = &input[to_absorb..];
let mut suboutput = Suboutout::security(self.bits);
unfinished.state.finalize(suboutput.as_bytes_mut());
self.state.update(suboutput.as_bytes());
self.blocks += 1;
} else {
unfinished.state.update(input);
unfinished.absorbed += input.len();
self.unfinished = Some(unfinished);
return;
}
}
let bits = self.bits;
let input_blocks_end = input.len() / self.block_size * self.block_size;
let input_blocks = &input[..input_blocks_end];
let input_end = &input[input_blocks_end..];
let parts = input_blocks.chunks(self.block_size).map(|chunk| {
let mut state = CShake::new(b"", b"", bits);
state.update(chunk);
let mut suboutput = Suboutout::security(bits);
state.finalize(suboutput.as_bytes_mut());
suboutput
});
for part in parts {
self.state.update(part.as_bytes());
self.blocks += 1;
}
if !input_end.is_empty() {
assert!(self.unfinished.is_none());
let mut state = CShake::new(b"", b"", bits);
state.update(input_end);
self.unfinished = Some(UnfinishedState {
state,
absorbed: input_end.len(),
});
}
}
fn finalize(mut self, output: &mut [u8]) {
if let Some(unfinished) = self.unfinished.take() {
let mut suboutput = Suboutout::security(self.bits);
unfinished.state.finalize(suboutput.as_bytes_mut());
self.state.update(suboutput.as_bytes());
self.blocks += 1;
}
self.state.update(right_encode(self.blocks).value());
self.state.update(right_encode(output.len() * 8).value());
self.state.finalize(output);
}
}
/// The `ParallelHashXOF` extendable-output functions defined in [`SP800-185`].
///
/// # Usage
///
/// ```toml
/// [dependencies]
/// tiny-keccak = { version = "2.0.0", features = ["parallel_hash"] }
/// ```
///
/// # Example
///
/// ```
/// # use tiny_keccak::{ParallelHash, Xof, IntoXof, Hasher};
/// let input = b"hello world";
/// let mut output = [0u8; 64];
/// let mut hasher = ParallelHash::v256(b"", 8);
/// hasher.update(input);
/// let mut xof = hasher.into_xof();
/// xof.squeeze(&mut output[..32]);
/// xof.squeeze(&mut output[32..]);
/// ```
///
/// ---
///
/// [`ParallelHashXof`] can be created only by using [`ParallelHash::IntoXof`] interface.
///
///
/// [`SP800-185`]: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-185.pdf
/// [`ParallelHashXof`]: struct.ParallelHashXof.html
/// [`ParallelHash::IntoXof`]: struct.ParallelHash.html#impl-IntoXof
#[derive(Clone)]
pub struct ParallelHashXof {
state: CShake,
}
impl IntoXof for ParallelHash {
type Xof = ParallelHashXof;
fn into_xof(mut self) -> Self::Xof {
if let Some(unfinished) = self.unfinished.take() {
let mut suboutput = Suboutout::security(self.bits);
unfinished.state.finalize(suboutput.as_bytes_mut());
self.state.update(suboutput.as_bytes());
self.blocks += 1;
}
self.state.update(right_encode(self.blocks).value());
self.state.update(right_encode(0).value());
ParallelHashXof { state: self.state }
}
}
impl Xof for ParallelHashXof {
fn squeeze(&mut self, output: &mut [u8]) {
self.state.squeeze(output);
}
}