executor: rewrite the work-stealing thread pool (#1657)

This patch is a ground up rewrite of the existing work-stealing thread
pool. The goal is to reduce overhead while simplifying code when
possible.

At a high level, the following architectural changes were made:

- The local run queues were switched for bounded circle buffer queues.
- Reduce cross-thread synchronization.
- Refactor task constructs to use a single allocation and always include
  a join handle (#887).
- Simplify logic around putting workers to sleep and waking them up.

**Local run queues**

Move away from crossbeam's implementation of the Chase-Lev deque. This
implementation included unnecessary overhead as it supported
capabilities that are not needed for the work-stealing thread pool.
Instead, a fixed size circle buffer is used for the local queue. When
the local queue is full, half of the tasks contained in it are moved to
the global run queue.

**Reduce cross-thread synchronization**

This is done via many small improvements. Primarily, an upper bound is
placed on the number of concurrent stealers. Limiting the number of
stealers results in lower contention. Secondly, the rate at which
workers are notified and woken up is throttled. This also reduces
contention by preventing many threads from racing to steal work.

**Refactor task structure**

Now that Tokio is able to target a rust version that supports
`std::alloc` as well as `std::task`, the pool is able to optimize how
the task structure is laid out. Now, a single allocation per task is
required and a join handle is always provided enabling the spawner to
retrieve the result of the task (#887).

**Simplifying logic**

When possible, complexity is reduced in the implementation. This is done
by using locks and other simpler constructs in cold paths. The set of
sleeping workers is now represented as a `Mutex<VecDeque<usize>>`.
Instead of optimizing access to this structure, we reduce the amount the
pool must access this structure.

Secondly, we have (temporarily) removed `threadpool::blocking`. This
capability will come back later, but the original implementation was way
more complicated than necessary.

**Results**

The thread pool benchmarks have improved significantly:

Old thread pool:

```
test chained_spawn ... bench:   2,019,796 ns/iter (+/- 302,168)
test ping_pong     ... bench:   1,279,948 ns/iter (+/- 154,365)
test spawn_many    ... bench:  10,283,608 ns/iter (+/- 1,284,275)
test yield_many    ... bench:  21,450,748 ns/iter (+/- 1,201,337)
```

New thread pool:

```
test chained_spawn ... bench:     147,943 ns/iter (+/- 6,673)
test ping_pong     ... bench:     537,744 ns/iter (+/- 20,928)
test spawn_many    ... bench:   7,454,898 ns/iter (+/- 283,449)
test yield_many    ... bench:  16,771,113 ns/iter (+/- 733,424)
```

Real-world benchmarks improve significantly as well. This is testing the hyper hello
world server using: `wrk -t1 -c50 -d10`:

Old scheduler:

```
Running 10s test @ http://127.0.0.1:3000
  1 threads and 50 connections
  Thread Stats   Avg      Stdev     Max   +/- Stdev
    Latency   371.53us   99.05us   1.97ms   60.53%
    Req/Sec   114.61k     8.45k  133.85k    67.00%
  1139307 requests in 10.00s, 95.61MB read
Requests/sec: 113923.19
Transfer/sec:      9.56MB
```

New scheduler:

```
Running 10s test @ http://127.0.0.1:3000
  1 threads and 50 connections
  Thread Stats   Avg      Stdev     Max   +/- Stdev
    Latency   275.05us   69.81us   1.09ms   73.57%
    Req/Sec   153.17k    10.68k  171.51k    71.00%
  1522671 requests in 10.00s, 127.79MB read
Requests/sec: 152258.70
Transfer/sec:     12.78MB
```

tokio-executor/src/util/mod.rs

@@ -0,0 +1,5 @@
+mod pad;
+mod rand;
+
+pub(crate) use self::pad::CachePadded;
+pub(crate) use self::rand::FastRand;

tokio-executor/src/util/pad.rs

@@ -0,0 +1,52 @@
+use core::fmt;
+use core::ops::{Deref, DerefMut};
+
+#[derive(Clone, Copy, Default, Hash, PartialEq, Eq)]
+// Starting from Intel's Sandy Bridge, spatial prefetcher is now pulling pairs of 64-byte cache
+// lines at a time, so we have to align to 128 bytes rather than 64.
+//
+// Sources:
+// - https://www.intel.com/content/dam/www/public/us/en/documents/manuals/64-ia-32-architectures-optimization-manual.pdf
+// - https://github.com/facebook/folly/blob/1b5288e6eea6df074758f877c849b6e73bbb9fbb/folly/lang/Align.h#L107
+#[cfg_attr(target_arch = "x86_64", repr(align(128)))]
+#[cfg_attr(not(target_arch = "x86_64"), repr(align(64)))]
+pub(crate) struct CachePadded<T> {
+    value: T,
+}
+
+unsafe impl<T: Send> Send for CachePadded<T> {}
+unsafe impl<T: Sync> Sync for CachePadded<T> {}
+
+impl<T> CachePadded<T> {
+    pub(crate) fn new(t: T) -> CachePadded<T> {
+        CachePadded::<T> { value: t }
+    }
+}
+
+impl<T> Deref for CachePadded<T> {
+    type Target = T;
+
+    fn deref(&self) -> &T {
+        &self.value
+    }
+}
+
+impl<T> DerefMut for CachePadded<T> {
+    fn deref_mut(&mut self) -> &mut T {
+        &mut self.value
+    }
+}
+
+impl<T: fmt::Debug> fmt::Debug for CachePadded<T> {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        f.debug_struct("CachePadded")
+            .field("value", &self.value)
+            .finish()
+    }
+}
+
+impl<T> From<T> for CachePadded<T> {
+    fn from(t: T) -> Self {
+        CachePadded::new(t)
+    }
+}

tokio-executor/src/util/rand.rs

@@ -0,0 +1,52 @@
+use std::cell::Cell;
+
+/// Fast random number generate
+///
+/// Implement xorshift64+: 2 32-bit xorshift sequences added together.
+/// Shift triplet [17,7,16] was calculated as indicated in Marsaglia's
+/// Xorshift paper: https://www.jstatsoft.org/article/view/v008i14/xorshift.pdf
+/// This generator passes the SmallCrush suite, part of TestU01 framework:
+/// http://simul.iro.umontreal.ca/testu01/tu01.html
+#[derive(Debug)]
+pub(crate) struct FastRand {
+    one: Cell<u32>,
+    two: Cell<u32>,
+}
+
+impl FastRand {
+    /// Initialize a new, thread-local, fast random number generator.
+    pub(crate) fn new(seed: u64) -> FastRand {
+        let one = (seed >> 32) as u32;
+        let mut two = seed as u32;
+
+        if two == 0 {
+            // This value cannot be zero
+            two = 1;
+        }
+
+        FastRand {
+            one: Cell::new(one),
+            two: Cell::new(two),
+        }
+    }
+
+    pub(crate) fn fastrand_n(&self, n: u32) -> u32 {
+        // This is similar to fastrand() % n, but faster.
+        // See https://lemire.me/blog/2016/06/27/a-fast-alternative-to-the-modulo-reduction/
+        let mul = (self.fastrand() as u64).wrapping_mul(n as u64);
+        (mul >> 32) as u32
+    }
+
+    fn fastrand(&self) -> u32 {
+        let mut s1 = self.one.get();
+        let s0 = self.two.get();
+
+        s1 ^= s1 << 17;
+        s1 = s1 ^ s0 ^ s1 >> 7 ^ s0 >> 16;
+
+        self.one.set(s0);
+        self.two.set(s1);
+
+        s0.wrapping_add(s1)
+    }
+}