core/hash/
sip.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
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
//! An implementation of SipHash.

#![allow(deprecated)] // the types in this module are deprecated

use crate::marker::PhantomData;
use crate::{cmp, mem, ptr};

/// An implementation of SipHash 1-3.
///
/// This is currently the default hashing function used by standard library
/// (e.g., `collections::HashMap` uses it by default).
///
/// See: <https://131002.net/siphash>
#[unstable(feature = "hashmap_internals", issue = "none")]
#[deprecated(since = "1.13.0", note = "use `std::hash::DefaultHasher` instead")]
#[derive(Debug, Clone, Default)]
#[doc(hidden)]
pub struct SipHasher13 {
    hasher: Hasher<Sip13Rounds>,
}

/// An implementation of SipHash 2-4.
///
/// See: <https://131002.net/siphash/>
#[unstable(feature = "hashmap_internals", issue = "none")]
#[deprecated(since = "1.13.0", note = "use `std::hash::DefaultHasher` instead")]
#[derive(Debug, Clone, Default)]
struct SipHasher24 {
    hasher: Hasher<Sip24Rounds>,
}

/// An implementation of SipHash 2-4.
///
/// See: <https://131002.net/siphash/>
///
/// SipHash is a general-purpose hashing function: it runs at a good
/// speed (competitive with Spooky and City) and permits strong _keyed_
/// hashing. This lets you key your hash tables from a strong RNG, such as
/// [`rand::os::OsRng`](https://docs.rs/rand/latest/rand/rngs/struct.OsRng.html).
///
/// Although the SipHash algorithm is considered to be generally strong,
/// it is not intended for cryptographic purposes. As such, all
/// cryptographic uses of this implementation are _strongly discouraged_.
#[stable(feature = "rust1", since = "1.0.0")]
#[deprecated(since = "1.13.0", note = "use `std::hash::DefaultHasher` instead")]
#[derive(Debug, Clone, Default)]
pub struct SipHasher(SipHasher24);

#[derive(Debug)]
struct Hasher<S: Sip> {
    k0: u64,
    k1: u64,
    length: usize, // how many bytes we've processed
    state: State,  // hash State
    tail: u64,     // unprocessed bytes le
    ntail: usize,  // how many bytes in tail are valid
    _marker: PhantomData<S>,
}

#[derive(Debug, Clone, Copy)]
#[repr(C)]
struct State {
    // v0, v2 and v1, v3 show up in pairs in the algorithm,
    // and simd implementations of SipHash will use vectors
    // of v02 and v13. By placing them in this order in the struct,
    // the compiler can pick up on just a few simd optimizations by itself.
    v0: u64,
    v2: u64,
    v1: u64,
    v3: u64,
}

macro_rules! compress {
    ($state:expr) => {{ compress!($state.v0, $state.v1, $state.v2, $state.v3) }};
    ($v0:expr, $v1:expr, $v2:expr, $v3:expr) => {{
        $v0 = $v0.wrapping_add($v1);
        $v2 = $v2.wrapping_add($v3);
        $v1 = $v1.rotate_left(13);
        $v1 ^= $v0;
        $v3 = $v3.rotate_left(16);
        $v3 ^= $v2;
        $v0 = $v0.rotate_left(32);

        $v2 = $v2.wrapping_add($v1);
        $v0 = $v0.wrapping_add($v3);
        $v1 = $v1.rotate_left(17);
        $v1 ^= $v2;
        $v3 = $v3.rotate_left(21);
        $v3 ^= $v0;
        $v2 = $v2.rotate_left(32);
    }};
}

/// Loads an integer of the desired type from a byte stream, in LE order. Uses
/// `copy_nonoverlapping` to let the compiler generate the most efficient way
/// to load it from a possibly unaligned address.
///
/// Safety: this performs unchecked indexing of `$buf` at
/// `$i..$i+size_of::<$int_ty>()`, so that must be in-bounds.
macro_rules! load_int_le {
    ($buf:expr, $i:expr, $int_ty:ident) => {{
        debug_assert!($i + mem::size_of::<$int_ty>() <= $buf.len());
        let mut data = 0 as $int_ty;
        ptr::copy_nonoverlapping(
            $buf.as_ptr().add($i),
            &mut data as *mut _ as *mut u8,
            mem::size_of::<$int_ty>(),
        );
        data.to_le()
    }};
}

/// Loads a u64 using up to 7 bytes of a byte slice. It looks clumsy but the
/// `copy_nonoverlapping` calls that occur (via `load_int_le!`) all have fixed
/// sizes and avoid calling `memcpy`, which is good for speed.
///
/// Safety: this performs unchecked indexing of `buf` at `start..start+len`, so
/// that must be in-bounds.
#[inline]
unsafe fn u8to64_le(buf: &[u8], start: usize, len: usize) -> u64 {
    debug_assert!(len < 8);
    let mut i = 0; // current byte index (from LSB) in the output u64
    let mut out = 0;
    if i + 3 < len {
        // SAFETY: `i` cannot be greater than `len`, and the caller must guarantee
        // that the index start..start+len is in bounds.
        out = unsafe { load_int_le!(buf, start + i, u32) } as u64;
        i += 4;
    }
    if i + 1 < len {
        // SAFETY: same as above.
        out |= (unsafe { load_int_le!(buf, start + i, u16) } as u64) << (i * 8);
        i += 2
    }
    if i < len {
        // SAFETY: same as above.
        out |= (unsafe { *buf.get_unchecked(start + i) } as u64) << (i * 8);
        i += 1;
    }
    //FIXME(fee1-dead): use debug_assert_eq
    debug_assert!(i == len);
    out
}

impl SipHasher {
    /// Creates a new `SipHasher` with the two initial keys set to 0.
    #[inline]
    #[stable(feature = "rust1", since = "1.0.0")]
    #[deprecated(since = "1.13.0", note = "use `std::hash::DefaultHasher` instead")]
    #[rustc_const_unstable(feature = "const_hash", issue = "104061")]
    #[must_use]
    pub const fn new() -> SipHasher {
        SipHasher::new_with_keys(0, 0)
    }

    /// Creates a `SipHasher` that is keyed off the provided keys.
    #[inline]
    #[stable(feature = "rust1", since = "1.0.0")]
    #[deprecated(since = "1.13.0", note = "use `std::hash::DefaultHasher` instead")]
    #[rustc_const_unstable(feature = "const_hash", issue = "104061")]
    #[must_use]
    pub const fn new_with_keys(key0: u64, key1: u64) -> SipHasher {
        SipHasher(SipHasher24 { hasher: Hasher::new_with_keys(key0, key1) })
    }
}

impl SipHasher13 {
    /// Creates a new `SipHasher13` with the two initial keys set to 0.
    #[inline]
    #[unstable(feature = "hashmap_internals", issue = "none")]
    #[deprecated(since = "1.13.0", note = "use `std::hash::DefaultHasher` instead")]
    #[rustc_const_unstable(feature = "const_hash", issue = "104061")]
    pub const fn new() -> SipHasher13 {
        SipHasher13::new_with_keys(0, 0)
    }

    /// Creates a `SipHasher13` that is keyed off the provided keys.
    #[inline]
    #[unstable(feature = "hashmap_internals", issue = "none")]
    #[deprecated(since = "1.13.0", note = "use `std::hash::DefaultHasher` instead")]
    #[rustc_const_unstable(feature = "const_hash", issue = "104061")]
    pub const fn new_with_keys(key0: u64, key1: u64) -> SipHasher13 {
        SipHasher13 { hasher: Hasher::new_with_keys(key0, key1) }
    }
}

impl<S: Sip> Hasher<S> {
    #[inline]
    const fn new_with_keys(key0: u64, key1: u64) -> Hasher<S> {
        let mut state = Hasher {
            k0: key0,
            k1: key1,
            length: 0,
            state: State { v0: 0, v1: 0, v2: 0, v3: 0 },
            tail: 0,
            ntail: 0,
            _marker: PhantomData,
        };
        state.reset();
        state
    }

    #[inline]
    const fn reset(&mut self) {
        self.length = 0;
        self.state.v0 = self.k0 ^ 0x736f6d6570736575;
        self.state.v1 = self.k1 ^ 0x646f72616e646f6d;
        self.state.v2 = self.k0 ^ 0x6c7967656e657261;
        self.state.v3 = self.k1 ^ 0x7465646279746573;
        self.ntail = 0;
    }
}

#[stable(feature = "rust1", since = "1.0.0")]
impl super::Hasher for SipHasher {
    #[inline]
    fn write(&mut self, msg: &[u8]) {
        self.0.hasher.write(msg)
    }

    #[inline]
    fn write_str(&mut self, s: &str) {
        self.0.hasher.write_str(s);
    }

    #[inline]
    fn finish(&self) -> u64 {
        self.0.hasher.finish()
    }
}

#[unstable(feature = "hashmap_internals", issue = "none")]
impl super::Hasher for SipHasher13 {
    #[inline]
    fn write(&mut self, msg: &[u8]) {
        self.hasher.write(msg)
    }

    #[inline]
    fn write_str(&mut self, s: &str) {
        self.hasher.write_str(s);
    }

    #[inline]
    fn finish(&self) -> u64 {
        self.hasher.finish()
    }
}

impl<S: Sip> super::Hasher for Hasher<S> {
    // Note: no integer hashing methods (`write_u*`, `write_i*`) are defined
    // for this type. We could add them, copy the `short_write` implementation
    // in librustc_data_structures/sip128.rs, and add `write_u*`/`write_i*`
    // methods to `SipHasher`, `SipHasher13`, and `DefaultHasher`. This would
    // greatly speed up integer hashing by those hashers, at the cost of
    // slightly slowing down compile speeds on some benchmarks. See #69152 for
    // details.
    #[inline]
    fn write(&mut self, msg: &[u8]) {
        let length = msg.len();
        self.length += length;

        let mut needed = 0;

        if self.ntail != 0 {
            needed = 8 - self.ntail;
            // SAFETY: `cmp::min(length, needed)` is guaranteed to not be over `length`
            self.tail |= unsafe { u8to64_le(msg, 0, cmp::min(length, needed)) } << (8 * self.ntail);
            if length < needed {
                self.ntail += length;
                return;
            } else {
                self.state.v3 ^= self.tail;
                S::c_rounds(&mut self.state);
                self.state.v0 ^= self.tail;
                self.ntail = 0;
            }
        }

        // Buffered tail is now flushed, process new input.
        let len = length - needed;
        let left = len & 0x7; // len % 8

        let mut i = needed;
        while i < len - left {
            // SAFETY: because `len - left` is the biggest multiple of 8 under
            // `len`, and because `i` starts at `needed` where `len` is `length - needed`,
            // `i + 8` is guaranteed to be less than or equal to `length`.
            let mi = unsafe { load_int_le!(msg, i, u64) };

            self.state.v3 ^= mi;
            S::c_rounds(&mut self.state);
            self.state.v0 ^= mi;

            i += 8;
        }

        // SAFETY: `i` is now `needed + len.div_euclid(8) * 8`,
        // so `i + left` = `needed + len` = `length`, which is by
        // definition equal to `msg.len()`.
        self.tail = unsafe { u8to64_le(msg, i, left) };
        self.ntail = left;
    }

    #[inline]
    fn write_str(&mut self, s: &str) {
        // This hasher works byte-wise, and `0xFF` cannot show up in a `str`,
        // so just hashing the one extra byte is enough to be prefix-free.
        self.write(s.as_bytes());
        self.write_u8(0xFF);
    }

    #[inline]
    fn finish(&self) -> u64 {
        let mut state = self.state;

        let b: u64 = ((self.length as u64 & 0xff) << 56) | self.tail;

        state.v3 ^= b;
        S::c_rounds(&mut state);
        state.v0 ^= b;

        state.v2 ^= 0xff;
        S::d_rounds(&mut state);

        state.v0 ^ state.v1 ^ state.v2 ^ state.v3
    }
}

impl<S: Sip> Clone for Hasher<S> {
    #[inline]
    fn clone(&self) -> Hasher<S> {
        Hasher {
            k0: self.k0,
            k1: self.k1,
            length: self.length,
            state: self.state,
            tail: self.tail,
            ntail: self.ntail,
            _marker: self._marker,
        }
    }
}

impl<S: Sip> Default for Hasher<S> {
    /// Creates a `Hasher<S>` with the two initial keys set to 0.
    #[inline]
    fn default() -> Hasher<S> {
        Hasher::new_with_keys(0, 0)
    }
}

#[doc(hidden)]
trait Sip {
    fn c_rounds(_: &mut State);
    fn d_rounds(_: &mut State);
}

#[derive(Debug, Clone, Default)]
struct Sip13Rounds;

impl Sip for Sip13Rounds {
    #[inline]
    fn c_rounds(state: &mut State) {
        compress!(state);
    }

    #[inline]
    fn d_rounds(state: &mut State) {
        compress!(state);
        compress!(state);
        compress!(state);
    }
}

#[derive(Debug, Clone, Default)]
struct Sip24Rounds;

impl Sip for Sip24Rounds {
    #[inline]
    fn c_rounds(state: &mut State) {
        compress!(state);
        compress!(state);
    }

    #[inline]
    fn d_rounds(state: &mut State) {
        compress!(state);
        compress!(state);
        compress!(state);
        compress!(state);
    }
}