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
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
use core::{
    fmt,
    marker::PhantomData,
    sync::atomic::{AtomicUsize, Ordering},
};

use crate::{AtomicTagPtr, TagPtr};

/********** impl Send + Sync **********************************************************************/

unsafe impl<T, const N: usize> Send for AtomicTagPtr<T, N> {}
unsafe impl<T, const N: usize> Sync for AtomicTagPtr<T, N> {}

/********** impl inherent *************************************************************************/

impl<T, const N: usize> AtomicTagPtr<T, N> {
    doc_comment! {
        doc_tag_bits!(),
        pub const TAG_BITS: usize = N;
    }

    doc_comment! {
        doc_tag_mask!(),
        pub const TAG_MASK: usize = crate::mark_mask(Self::TAG_BITS);
    }

    doc_comment! {
        doc_ptr_mask!(),
        pub const POINTER_MASK: usize = !Self::TAG_MASK;
    }

    doc_comment! {
        doc_null!(),
        ///
        /// # Examples
        ///
        /// ```
        /// use core::{ptr, sync::atomic::Ordering};
        ///
        /// type AtomicTagPtr = tagptr::AtomicTagPtr<i32, 2>;
        ///
        /// let ptr = AtomicTagPtr::null();
        /// assert_eq!(
        ///     ptr.load(Ordering::Relaxed).decompose(),
        ///     (ptr::null_mut(), 0)
        /// );
        /// ```
        pub const fn null() -> Self {
            Self { inner: AtomicUsize::new(0), _marker: PhantomData }
        }
    }

    doc_comment! {
        doc_atomic_new!(),
        #[inline]
        pub fn new(marked_ptr: TagPtr<T, N>) -> Self {
            Self { inner: AtomicUsize::new(marked_ptr.into_usize()), _marker: PhantomData }
        }
    }

    doc_comment! {
        doc_atomic_into_inner!(),
        #[inline]
        pub fn into_inner(self) -> TagPtr<T, N> {
            TagPtr::from_usize(self.inner.into_inner())
        }
    }

    /// Returns a mutable reference to the underlying marked pointer.
    ///
    /// This is safe because the mutable reference guarantees no other
    /// threads are concurrently accessing the atomic pointer.
    #[inline]
    pub fn get_mut(&mut self) -> &mut TagPtr<T, N> {
        // SAFETY: the mutable self reference ensures the dereferencing is sound
        unsafe { &mut *(self.inner.get_mut() as *mut usize as *mut _) }
    }

    /// Loads the value of the atomic marked pointer.
    ///
    /// `load` takes an [`Ordering`] argument which describes the memory
    /// ordering of this operation.
    /// Possible values are [`SeqCst`][seq_cst], [`Acquire`][acq] and
    /// [`Relaxed`][rlx].
    ///
    /// # Panics
    ///
    /// Panics if `order` is [`Release`][rel] or [`AcqRel`][acq_rel].
    ///
    /// [rlx]: Ordering::Relaxed
    /// [acq]: Ordering::Acquire
    /// [rel]: Ordering::Release
    /// [acq_rel]: Ordering::AcqRel
    /// [seq_cst]: Ordering::SeqCst
    #[inline]
    pub fn load(&self, order: Ordering) -> TagPtr<T, N> {
        TagPtr::from_usize(self.inner.load(order))
    }

    /// Stores a value into the atomic marked pointer.
    ///
    /// `store` takes an [`Ordering`] argument which describes the memory
    /// ordering of this operation.
    /// Possible values are [`SeqCst`][seq_cst], [`Release`][rel] and
    /// [`Relaxed`][rlx].
    ///
    /// # Panics
    ///
    /// Panics if `order` is [`Acquire`][acq] or [`AcqRel`][acq_rel].
    ///
    /// [rlx]: Ordering::Relaxed
    /// [acq]: Ordering::Acquire
    /// [rel]: Ordering::Release
    /// [acq_rel]: Ordering::AcqRel
    /// [seq_cst]: Ordering::SeqCst
    #[inline]
    pub fn store(&self, ptr: TagPtr<T, N>, order: Ordering) {
        self.inner.store(ptr.into_usize(), order)
    }

    /// Stores a value into the atomic marked pointer and returns the previous
    /// value.
    ///
    /// `swap` takes an [`Ordering`] argument which describes the memory
    /// ordering of this operation.
    /// All ordering modes are possible.
    /// Note that using [`Acquire`][acq] makes the store part of this operation
    /// [`Relaxed`][rlx], and using [`Release`][rel] makes the load part
    /// [`Relaxed`][rlx].
    ///
    /// [rlx]: Ordering::Relaxed
    /// [acq]: Ordering::Acquire
    /// [rel]: Ordering::Release
    ///
    /// # Examples
    ///
    /// ```
    /// use core::sync::atomic::Ordering;
    ///
    /// type AtomicTagPtr = tagptr::AtomicTagPtr<i32, 2>;
    /// type TagPtr = tagptr::TagPtr<i32, 2>;
    ///
    /// let ptr = AtomicTagPtr::null();
    /// let prev = ptr.swap(TagPtr::new(&mut 1), Ordering::Relaxed);
    ///
    /// assert!(prev.is_null());
    /// ```
    pub fn swap(&self, ptr: TagPtr<T, N>, order: Ordering) -> TagPtr<T, N> {
        TagPtr::from_usize(self.inner.swap(ptr.into_usize(), order))
    }

    /// Stores a value into the pointer if the current value is the same as
    /// `current`.
    ///
    /// The return value is a result indicating whether the new value was
    /// written and containing the previous value.
    /// On success this value is guaranteed to be equal to `current`.
    ///
    /// `compare_exchange` takes takes two [`Ordering`] arguments to describe
    /// the memory ordering of this operation.
    /// The first describes the required ordering if the operation succeeds
    /// while the second describes the required ordering when the operation
    /// fails.
    /// Using [`Acquire`][acq] as success ordering makes store part of this
    /// operation [`Relaxed`][rlx], and using [`Release`][rel] makes the
    /// successful load [`Relaxed`][rlx].
    /// The failure ordering can only be [`SeqCst`][seq_cst], [`Acquire`][acq]
    /// or [`Relaxed`][rlx] and must be equivalent or weaker than the success
    /// ordering.
    ///
    /// [rlx]: Ordering::Relaxed
    /// [acq]: Ordering::Acquire
    /// [rel]: Ordering::Release
    /// [seq_cst]: Ordering::SeqCst
    #[inline]
    pub fn compare_exchange(
        &self,
        current: TagPtr<T, N>,
        new: TagPtr<T, N>,
        (success, failure): (Ordering, Ordering),
    ) -> Result<TagPtr<T, N>, TagPtr<T, N>> {
        self.inner
            .compare_exchange(current.into_usize(), new.into_usize(), success, failure)
            .map(|_| current)
            .map_err(TagPtr::from_usize)
    }

    /// Stores a value into the pointer if the current value is the same as
    /// `current`.
    ///
    /// The return value is a result indicating whether the new value was
    /// written and containing the previous value.
    /// On success this value is guaranteed to be equal to `current`.
    ///
    /// Unlike `compare_exchange`, this function is allowed to spuriously fail,
    /// even when the comparison succeeds, which can result in more efficient
    /// code on some platforms.
    /// The return value is a result indicating whether the new value was
    /// written and containing the previous value.
    ///
    /// `compare_exchange` takes takes two [`Ordering`] arguments to describe
    /// the memory ordering of this operation.
    /// The first describes the required ordering if the operation succeeds
    /// while the second describes the required ordering when the operation
    /// fails.
    /// Using [`Acquire`][acq] as success ordering makes store part of this
    /// operation [`Relaxed`][rlx], and using [`Release`][rel] makes the
    /// successful load [`Relaxed`][rlx].
    /// The failure ordering can only be [`SeqCst`][seq_cst], [`Acquire`][acq]
    /// or [`Relaxed`][rlx] and must be equivalent or weaker than the success
    /// ordering.
    ///
    /// [rlx]: Ordering::Relaxed
    /// [acq]: Ordering::Acquire
    /// [rel]: Ordering::Release
    /// [seq_cst]: Ordering::SeqCst
    #[inline]
    pub fn compare_exchange_weak(
        &self,
        current: TagPtr<T, N>,
        new: TagPtr<T, N>,
        (success, failure): (Ordering, Ordering),
    ) -> Result<TagPtr<T, N>, TagPtr<T, N>> {
        self.inner
            .compare_exchange_weak(current.into_usize(), new.into_usize(), success, failure)
            .map(|_| current)
            .map_err(TagPtr::from_usize)
    }

    /// Adds `value` to the current tag value, returning the previous marked
    /// pointer.
    ///
    /// This operation directly and unconditionally alters the internal numeric
    /// representation of the atomic marked pointer.
    /// Hence there is no way to reliably guarantee the operation only affects
    /// the tag bits and does not overflow into the pointer bits.
    ///
    /// `fetch_add` takes takes an [`Ordering`] argument which describes the
    /// memory ordering of this operation.
    /// All ordering modes are possible.
    /// Note that using [`Acquire`][acq] makes the store part of this operation
    /// [`Relaxed`][rlx] and using [`Release`][rel] makes the load part
    /// [`Relaxed`][rlx].
    ///
    /// [rlx]: Ordering::Relaxed
    /// [acq]: Ordering::Acquire
    /// [rel]: Ordering::Release
    ///
    /// # Examples
    ///
    /// ```
    /// use core::sync::atomic::Ordering;
    ///
    /// type AtomicTagPtr = tagptr::AtomicTagPtr<i32, 2>;
    /// type TagPtr = tagptr::TagPtr<i32, 2>;
    ///
    /// let reference = &mut 1;
    /// let ptr = AtomicTagPtr::new(TagPtr::new(reference));
    ///
    /// assert_eq!(
    ///     ptr.fetch_add(1, Ordering::Relaxed).decompose(),
    ///     (reference as *mut _, 0)
    /// );
    ///
    /// assert_eq!(
    ///     ptr.load(Ordering::Relaxed).decompose(),
    ///     (reference as *mut _, 0b01)
    /// );
    /// ```
    #[inline]
    pub fn fetch_add(&self, value: usize, order: Ordering) -> TagPtr<T, N> {
        debug_assert!(value < Self::TAG_MASK, "`value` exceeds tag bits (would overflow)");
        TagPtr::from_usize(self.inner.fetch_add(value, order))
    }

    /// Subtracts `value` from the current tag value, returning the previous
    /// marked pointer.
    ///
    /// This operation directly and unconditionally alters the internal numeric
    /// representation of the atomic marked pointer.
    /// Hence there is no way to reliably guarantee the operation only affects
    /// the tag bits and does not overflow into the pointer bits.
    ///
    /// `fetch_sub` takes takes an [`Ordering`] argument which describes the
    /// memory ordering of this operation.
    /// All ordering modes are possible.
    /// Note that using [`Acquire`][acq] makes the store part of this operation
    /// [`Relaxed`][rlx] and using [`Release`][rel] makes the load part
    /// [`Relaxed`][rlx].
    ///
    /// [rlx]: Ordering::Relaxed
    /// [acq]: Ordering::Acquire
    /// [rel]: Ordering::Release
    ///
    /// # Examples
    ///
    /// ```
    /// use core::sync::atomic::Ordering;
    ///
    /// type AtomicTagPtr = tagptr::AtomicTagPtr<i32, 2>;
    /// type TagPtr = tagptr::TagPtr<i32, 2>;
    ///
    /// let reference = &mut 1;
    /// let ptr = AtomicTagPtr::new(TagPtr::compose(reference, 0b10));
    ///
    /// assert_eq!(
    ///     ptr.fetch_sub(1, Ordering::Relaxed).decompose(),
    ///     (reference as *mut _, 0b10)
    /// );
    ///
    /// assert_eq!(
    ///     ptr.load(Ordering::Relaxed).decompose(),
    ///     (reference as *mut _, 0b01)
    /// );
    /// ```
    #[inline]
    pub fn fetch_sub(&self, value: usize, order: Ordering) -> TagPtr<T, N> {
        debug_assert!(value < Self::TAG_MASK, "`value` exceeds tag bits (would underflow)");
        TagPtr::from_usize(self.inner.fetch_sub(value, order))
    }

    /// Performs a bitwise "or" of `value` with the current tag value, returning
    /// the previous marked pointer.
    ///
    /// This operation directly and unconditionally alters the internal numeric
    /// representation of the atomic marked pointer.
    /// Hence there is no way to reliably guarantee the operation only affects
    /// the tag bits and does not overflow into the pointer bits.
    ///
    /// `fetch_or` takes takes an [`Ordering`] argument which describes the
    /// memory ordering of this operation.
    /// All ordering modes are possible.
    /// Note that using [`Acquire`][acq] makes the store part of this operation
    /// [`Relaxed`][rlx] and using [`Release`][rel] makes the load part
    /// [`Relaxed`][rlx].
    ///
    /// [rlx]: Ordering::Relaxed
    /// [acq]: Ordering::Acquire
    /// [rel]: Ordering::Release
    ///
    /// # Examples
    ///
    /// ```
    /// use core::sync::atomic::Ordering;
    ///
    /// type AtomicTagPtr = tagptr::AtomicTagPtr<i32, 2>;
    /// type TagPtr = tagptr::TagPtr<i32, 2>;
    ///
    /// let reference = &mut 1;
    /// let ptr = AtomicTagPtr::new(TagPtr::compose(reference, 0b10));
    ///
    /// assert_eq!(
    ///     ptr.fetch_or(0b11, Ordering::Relaxed).decompose(),
    ///     (reference as *mut _, 0b10)
    /// );
    ///
    /// assert_eq!(
    ///     ptr.load(Ordering::Relaxed).decompose(),
    ///     (reference as *mut _, 0b11)
    /// );
    /// ```
    #[inline]
    pub fn fetch_or(&self, value: usize, order: Ordering) -> TagPtr<T, N> {
        debug_assert!(value <= Self::TAG_MASK, "`value` exceeds tag bits (would corrupt pointer)");
        TagPtr::from_usize(self.inner.fetch_or(Self::TAG_MASK & value, order))
    }

    /// Performs a bitwise "and" of `value` with the current tag value,
    /// returning the previous marked pointer.
    ///
    /// This operation directly and unconditionally alters the internal numeric
    /// representation of the atomic marked pointer.
    /// Hence there is no way to reliably guarantee the operation only affects
    /// the tag bits and does not overflow into the pointer bits.
    ///
    /// `fetch_and` takes takes an [`Ordering`] argument which describes the
    /// memory ordering of this operation.
    /// All ordering modes are possible.
    /// Note that using [`Acquire`][acq] makes the store part of this operation
    /// [`Relaxed`][rlx] and using [`Release`][rel] makes the load part
    /// [`Relaxed`][rlx].
    ///
    /// [rlx]: Ordering::Relaxed
    /// [acq]: Ordering::Acquire
    /// [rel]: Ordering::Release
    ///
    /// # Examples
    ///
    /// ```
    /// use core::sync::atomic::Ordering;
    ///
    /// type AtomicTagPtr = tagptr::AtomicTagPtr<i32, 2>;
    /// type TagPtr = tagptr::TagPtr<i32, 2>;
    ///
    /// let reference = &mut 1;
    /// let ptr = AtomicTagPtr::new(TagPtr::compose(reference, 0b10));
    ///
    /// // fetch_x returns previous value
    /// assert_eq!(
    ///     ptr.fetch_and(0b11, Ordering::Relaxed).decompose(),
    ///     (reference as *mut _, 0b10)
    /// );
    ///
    /// assert_eq!(
    ///     ptr.load(Ordering::Relaxed).decompose(),
    ///     (reference as *mut _, 0b10)
    /// );
    /// ```
    #[inline]
    pub fn fetch_and(&self, value: usize, order: Ordering) -> TagPtr<T, N> {
        debug_assert!(value <= Self::TAG_MASK, "`value` exceeds tag bits (would corrupt pointer)");
        TagPtr::from_usize(self.inner.fetch_and(Self::POINTER_MASK | value, order))
    }
}

/********** impl Debug ****************************************************************************/

impl<T, const N: usize> fmt::Debug for AtomicTagPtr<T, N> {
    #[inline]
    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
        let (ptr, tag) = self.load(Ordering::SeqCst).decompose();
        f.debug_struct("AtomicTagPtr").field("ptr", &ptr).field("tag", &tag).finish()
    }
}

/********** impl Default **************************************************************************/

impl<T, const N: usize> Default for AtomicTagPtr<T, N> {
    impl_default!();
}

/********** impl From (*mut T) ********************************************************************/

impl<T, const N: usize> From<*mut T> for AtomicTagPtr<T, N> {
    #[inline]
    fn from(ptr: *mut T) -> Self {
        Self::new(ptr.into())
    }
}

/********** impl From (TagPtr<T, N>) ***********************************************************/

impl<T, const N: usize> From<TagPtr<T, N>> for AtomicTagPtr<T, N> {
    #[inline]
    fn from(ptr: TagPtr<T, N>) -> Self {
        Self::new(ptr)
    }
}

/********** impl Pointer **************************************************************************/

impl<T, const N: usize> fmt::Pointer for AtomicTagPtr<T, N> {
    #[inline]
    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
        fmt::Pointer::fmt(&self.load(Ordering::SeqCst), f)
    }
}