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
use crate::loom::cell::{CausalCell, CausalCheck};
use crate::loom::sync::atomic::{self, AtomicU32};
use crate::runtime::thread_pool::queue::global;
use crate::runtime::thread_pool::LOCAL_QUEUE_CAPACITY;
use crate::task::Task;
use std::fmt;
use std::mem::MaybeUninit;
use std::ptr;
use std::sync::atomic::Ordering::{Acquire, Release};
pub(super) struct Queue<T: 'static> {
/// Concurrently updated by many threads.
head: AtomicU32,
/// Only updated by producer thread but read by many threads.
tail: AtomicU32,
/// Elements
buffer: Box<[CausalCell<MaybeUninit<Task<T>>>]>,
}
const MASK: usize = LOCAL_QUEUE_CAPACITY - 1;
impl<T: 'static> Queue<T> {
pub(super) fn new() -> Queue<T> {
debug_assert!(LOCAL_QUEUE_CAPACITY >= 2 && LOCAL_QUEUE_CAPACITY.is_power_of_two());
let mut buffer = Vec::with_capacity(LOCAL_QUEUE_CAPACITY);
for _ in 0..LOCAL_QUEUE_CAPACITY {
buffer.push(CausalCell::new(MaybeUninit::uninit()));
}
Queue {
head: AtomicU32::new(0),
tail: AtomicU32::new(0),
buffer: buffer.into(),
}
}
}
impl<T> Queue<T> {
/// Pushes a task onto the local queue.
///
/// This **must** be called by the producer thread.
pub(super) unsafe fn push(&self, mut task: Task<T>, global: &global::Queue<T>) {
loop {
let head = self.head.load(Acquire);
// safety: this is the **only** thread that updates this cell.
let tail = self.tail.unsync_load();
if tail.wrapping_sub(head) < LOCAL_QUEUE_CAPACITY as u32 {
// Map the position to a slot index.
let idx = tail as usize & MASK;
self.buffer[idx].with_mut(|ptr| {
// Write the task to the slot
ptr::write((*ptr).as_mut_ptr(), task);
});
// Make the task available
self.tail.store(tail.wrapping_add(1), Release);
return;
}
// The local buffer is full. Push a batch of work to the global
// queue.
match self.push_overflow(task, head, tail, global) {
Ok(_) => return,
// Lost the race, try again
Err(v) => task = v,
}
atomic::spin_loop_hint();
}
}
/// Moves a batch of tasks into the global queue.
///
/// This will temporarily make some of the tasks unavailable to stealers.
/// Once `push_overflow` is done, a notification is sent out, so if other
/// workers "missed" some of the tasks during a steal, they will get
/// another opportunity.
#[inline(never)]
unsafe fn push_overflow(
&self,
task: Task<T>,
head: u32,
tail: u32,
global: &global::Queue<T>,
) -> Result<(), Task<T>> {
const BATCH_LEN: usize = LOCAL_QUEUE_CAPACITY / 2 + 1;
let n = tail.wrapping_sub(head) / 2;
assert_eq!(n as usize, LOCAL_QUEUE_CAPACITY / 2, "queue is not full");
// Claim a bunch of tasks
//
// We are claiming the tasks **before** reading them out of the buffer.
// This is safe because only the **current** thread is able to push new
// tasks.
//
// There isn't really any need for memory ordering... Relaxed would
// work. This is because all tasks are pushed into the queue from the
// current thread (or memory has been acquired if the local queue handle
// moved).
let actual = self.head.compare_and_swap(head, head + n, Release);
if actual != head {
// We failed to claim the tasks, losing the race. Return out of
// this function and try the full `push` routine again. The queue
// may not be full anymore.
return Err(task);
}
// link the tasks
for i in 0..n {
let j = i + 1;
let i_idx = (i + head) as usize & MASK;
let j_idx = (j + head) as usize & MASK;
// Get the next pointer
let next = if j == n {
// The last task in the local queue being moved
task.header() as *const _
} else {
self.buffer[j_idx].with(|ptr| {
let value = (*ptr).as_ptr();
(*value).header() as *const _
})
};
self.buffer[i_idx].with_mut(|ptr| {
let ptr = (*ptr).as_ptr();
debug_assert!((*(*ptr).header().queue_next.get()).is_null());
*(*ptr).header().queue_next.get() = next;
});
}
let head = self.buffer[head as usize & MASK].with(|ptr| ptr::read((*ptr).as_ptr()));
// Push the tasks onto the global queue
global.push_batch(head, task, BATCH_LEN);
Ok(())
}
/// Pops a task from the local queue.
///
/// This **must** be called by the producer thread
pub(super) unsafe fn pop(&self) -> Option<Task<T>> {
loop {
let head = self.head.load(Acquire);
// safety: this is the **only** thread that updates this cell.
let tail = self.tail.unsync_load();
if head == tail {
// queue is empty
return None;
}
// Map the head position to a slot index.
let idx = head as usize & MASK;
let task = self.buffer[idx].with(|ptr| {
// Tentatively read the task at the head position. Note that we
// have not yet claimed the task.
//
ptr::read(ptr)
});
// Attempt to claim the task read above.
let actual = self
.head
.compare_and_swap(head, head.wrapping_add(1), Release);
if actual == head {
return Some(task.assume_init());
}
atomic::spin_loop_hint();
}
}
pub(super) fn is_empty(&self) -> bool {
let head = self.head.load(Acquire);
let tail = self.tail.load(Acquire);
head == tail
}
/// Steals half the tasks from self and place them into `dst`.
pub(super) unsafe fn steal(&self, dst: &Queue<T>) -> Option<Task<T>> {
let dst_tail = dst.tail.unsync_load();
// Steal the tasks into `dst`'s buffer. This does not yet expose the
// tasks in `dst`.
let mut n = self.steal2(dst, dst_tail);
if n == 0 {
// No tasks were stolen
return None;
}
// We are returning a task here
n -= 1;
let ret_pos = dst_tail.wrapping_add(n);
let ret_idx = ret_pos as usize & MASK;
let ret = dst.buffer[ret_idx].with(|ptr| ptr::read((*ptr).as_ptr()));
if n == 0 {
// The `dst` queue is empty, but a single task was stolen
return Some(ret);
}
// Synchronize with stealers
let dst_head = dst.head.load(Acquire);
assert!(dst_tail.wrapping_sub(dst_head) + n <= LOCAL_QUEUE_CAPACITY as u32);
// Make the stolen items available to consumers
dst.tail.store(dst_tail.wrapping_add(n), Release);
Some(ret)
}
unsafe fn steal2(&self, dst: &Queue<T>, dst_tail: u32) -> u32 {
loop {
let src_head = self.head.load(Acquire);
let src_tail = self.tail.load(Acquire);
// Number of available tasks to steal
let n = src_tail.wrapping_sub(src_head);
let n = n - n / 2;
if n == 0 {
return 0;
}
if n > LOCAL_QUEUE_CAPACITY as u32 / 2 {
atomic::spin_loop_hint();
// inconsistent, try again
continue;
}
// Track CausalCell causality checks. The check is deferred until
// the compare_and_swap claims ownership of the tasks.
let mut check = CausalCheck::default();
for i in 0..n {
// Compute the positions
let src_pos = src_head.wrapping_add(i);
let dst_pos = dst_tail.wrapping_add(i);
// Map to slots
let src_idx = src_pos as usize & MASK;
let dst_idx = dst_pos as usize & MASK;
// Read the task
let (task, ch) =
self.buffer[src_idx].with_deferred(|ptr| ptr::read((*ptr).as_ptr()));
check.join(ch);
// Write the task to the new slot
dst.buffer[dst_idx].with_mut(|ptr| ptr::write((*ptr).as_mut_ptr(), task));
}
// Claim all of those tasks!
let actual = self
.head
.compare_and_swap(src_head, src_head.wrapping_add(n), Release);
if actual == src_head {
check.check();
return n;
}
atomic::spin_loop_hint();
}
}
}
impl<T> fmt::Debug for Queue<T> {
fn fmt(&self, fmt: &mut fmt::Formatter<'_>) -> fmt::Result {
fmt.debug_struct("local::Queue")
.field("head", &self.head)
.field("tail", &self.tail)
.field("buffer", &"[...]")
.finish()
}
}