const SIGMA: [u32; 4] = [0x6170_7865, 0x3320_646e, 0x7962_2d32, 0x6b20_6574];
fn le32(b: &[u8]) -> u32 {
u32::from_le_bytes([b[0], b[1], b[2], b[3]])
}
fn quarter_round(s: &mut [u32; 16], a: usize, b: usize, c: usize, d: usize) {
s[a] = s[a].wrapping_add(s[b]);
s[d] = (s[d] ^ s[a]).rotate_left(16);
s[c] = s[c].wrapping_add(s[d]);
s[b] = (s[b] ^ s[c]).rotate_left(12);
s[a] = s[a].wrapping_add(s[b]);
s[d] = (s[d] ^ s[a]).rotate_left(8);
s[c] = s[c].wrapping_add(s[d]);
s[b] = (s[b] ^ s[c]).rotate_left(7);
}
pub(crate) fn block(key: &[u8; 32], counter: u32, nonce: &[u8; 12]) -> [u8; 64] {
let mut init = [0u32; 16];
init[..4].copy_from_slice(&SIGMA);
for i in 0..8 {
init[4 + i] = le32(&key[4 * i..]);
}
init[12] = counter;
for i in 0..3 {
init[13 + i] = le32(&nonce[4 * i..]);
}
let mut s = init;
for _ in 0..10 {
quarter_round(&mut s, 0, 4, 8, 12);
quarter_round(&mut s, 1, 5, 9, 13);
quarter_round(&mut s, 2, 6, 10, 14);
quarter_round(&mut s, 3, 7, 11, 15);
quarter_round(&mut s, 0, 5, 10, 15);
quarter_round(&mut s, 1, 6, 11, 12);
quarter_round(&mut s, 2, 7, 8, 13);
quarter_round(&mut s, 3, 4, 9, 14);
}
let mut out = [0u8; 64];
for i in 0..16 {
out[4 * i..4 * i + 4].copy_from_slice(&s[i].wrapping_add(init[i]).to_le_bytes());
}
out
}
pub(crate) fn xor_keystream(key: &[u8; 32], counter: u32, nonce: &[u8; 12], buf: &mut [u8]) {
let mut ctr = counter;
for chunk in buf.chunks_mut(64) {
let ks = block(key, ctr, nonce);
for (b, k) in chunk.iter_mut().zip(ks.iter()) {
*b ^= k;
}
ctr = ctr.wrapping_add(1);
}
}