max/embassy
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Split embassy-time from embassy-executor.

Browse Source
This commit is contained in:
Dario Nieuwenhuis
2022-08-17 23:40:16 +02:00
parent 1c5b54a482
commit 5daa173ce4
254 changed files with 847 additions and 552 deletions
Download Patch File Download Diff File Expand all files Collapse all files

428
embassy-executor/src/raw/mod.rs Normal file
View File

@ -0,0 +1,428 @@
//! Raw executor.
//!
//! This module exposes "raw" Executor and Task structs for more low level control.
//!
//! ## WARNING: here be dragons!
//!
//! Using this module requires respecting subtle safety contracts. If you can, prefer using the safe
//! executor wrappers in [`executor`](crate::executor) and the [`embassy_executor::task`](embassy_macros::task) macro, which are fully safe.
mod run_queue;
#[cfg(feature = "integrated-timers")]
mod timer_queue;
pub(crate) mod util;
mod waker;
use core::cell::Cell;
use core::future::Future;
use core::pin::Pin;
use core::ptr::NonNull;
use core::task::{Context, Poll};
use core::{mem, ptr};
use atomic_polyfill::{AtomicU32, Ordering};
use critical_section::CriticalSection;
#[cfg(feature = "integrated-timers")]
use embassy_time::driver::{self, AlarmHandle};
#[cfg(feature = "integrated-timers")]
use embassy_time::Instant;
use self::run_queue::{RunQueue, RunQueueItem};
use self::util::UninitCell;
pub use self::waker::task_from_waker;
use super::SpawnToken;
/// Task is spawned (has a future)
pub(crate) const STATE_SPAWNED: u32 = 1 << 0;
/// Task is in the executor run queue
pub(crate) const STATE_RUN_QUEUED: u32 = 1 << 1;
/// Task is in the executor timer queue
#[cfg(feature = "integrated-timers")]
pub(crate) const STATE_TIMER_QUEUED: u32 = 1 << 2;
/// Raw task header for use in task pointers.
///
/// This is an opaque struct, used for raw pointers to tasks, for use
/// with funtions like [`wake_task`] and [`task_from_waker`].
pub struct TaskHeader {
pub(crate) state: AtomicU32,
pub(crate) run_queue_item: RunQueueItem,
pub(crate) executor: Cell<*const Executor>, // Valid if state != 0
pub(crate) poll_fn: UninitCell<unsafe fn(NonNull<TaskHeader>)>, // Valid if STATE_SPAWNED
#[cfg(feature = "integrated-timers")]
pub(crate) expires_at: Cell<Instant>,
#[cfg(feature = "integrated-timers")]
pub(crate) timer_queue_item: timer_queue::TimerQueueItem,
}
impl TaskHeader {
pub(crate) const fn new() -> Self {
Self {
state: AtomicU32::new(0),
run_queue_item: RunQueueItem::new(),
executor: Cell::new(ptr::null()),
poll_fn: UninitCell::uninit(),
#[cfg(feature = "integrated-timers")]
expires_at: Cell::new(Instant::from_ticks(0)),
#[cfg(feature = "integrated-timers")]
timer_queue_item: timer_queue::TimerQueueItem::new(),
}
}
}
/// Raw storage in which a task can be spawned.
///
/// This struct holds the necessary memory to spawn one task whose future is `F`.
/// At a given time, the `TaskStorage` may be in spawned or not-spawned state. You
/// may spawn it with [`TaskStorage::spawn()`], which will fail if it is already spawned.
///
/// A `TaskStorage` must live forever, it may not be deallocated even after the task has finished
/// running. Hence the relevant methods require `&'static self`. It may be reused, however.
///
/// Internally, the [embassy_executor::task](embassy_macros::task) macro allocates an array of `TaskStorage`s
/// in a `static`. The most common reason to use the raw `Task` is to have control of where
/// the memory for the task is allocated: on the stack, or on the heap with e.g. `Box::leak`, etc.
// repr(C) is needed to guarantee that the Task is located at offset 0
// This makes it safe to cast between TaskHeader and TaskStorage pointers.
#[repr(C)]
pub struct TaskStorage<F: Future + 'static> {
raw: TaskHeader,
future: UninitCell<F>, // Valid if STATE_SPAWNED
}
impl<F: Future + 'static> TaskStorage<F> {
const NEW: Self = Self::new();
/// Create a new TaskStorage, in not-spawned state.
pub const fn new() -> Self {
Self {
raw: TaskHeader::new(),
future: UninitCell::uninit(),
}
}
/// Try to spawn the task.
///
/// The `future` closure constructs the future. It's only called if spawning is
/// actually possible. It is a closure instead of a simple `future: F` param to ensure
/// the future is constructed in-place, avoiding a temporary copy in the stack thanks to
/// NRVO optimizations.
///
/// This function will fail if the task is already spawned and has not finished running.
/// In this case, the error is delayed: a "poisoned" SpawnToken is returned, which will
/// cause [`Spawner::spawn()`](super::Spawner::spawn) to return the error.
///
/// Once the task has finished running, you may spawn it again. It is allowed to spawn it
/// on a different executor.
pub fn spawn(&'static self, future: impl FnOnce() -> F) -> SpawnToken<impl Sized> {
if self.spawn_mark_used() {
return unsafe { SpawnToken::<F>::new(self.spawn_initialize(future)) };
}
SpawnToken::<F>::new_failed()
}
fn spawn_mark_used(&'static self) -> bool {
let state = STATE_SPAWNED | STATE_RUN_QUEUED;
self.raw
.state
.compare_exchange(0, state, Ordering::AcqRel, Ordering::Acquire)
.is_ok()
}
unsafe fn spawn_initialize(&'static self, future: impl FnOnce() -> F) -> NonNull<TaskHeader> {
// Initialize the task
self.raw.poll_fn.write(Self::poll);
self.future.write(future());
NonNull::new_unchecked(self as *const TaskStorage<F> as *const TaskHeader as *mut TaskHeader)
}
unsafe fn poll(p: NonNull<TaskHeader>) {
let this = &*(p.as_ptr() as *const TaskStorage<F>);
let future = Pin::new_unchecked(this.future.as_mut());
let waker = waker::from_task(p);
let mut cx = Context::from_waker(&waker);
match future.poll(&mut cx) {
Poll::Ready(_) => {
this.future.drop_in_place();
this.raw.state.fetch_and(!STATE_SPAWNED, Ordering::AcqRel);
}
Poll::Pending => {}
}
// the compiler is emitting a virtual call for waker drop, but we know
// it's a noop for our waker.
mem::forget(waker);
}
}
unsafe impl<F: Future + 'static> Sync for TaskStorage<F> {}
/// Raw storage that can hold up to N tasks of the same type.
///
/// This is essentially a `[TaskStorage<F>; N]`.
pub struct TaskPool<F: Future + 'static, const N: usize> {
pool: [TaskStorage<F>; N],
}
impl<F: Future + 'static, const N: usize> TaskPool<F, N> {
/// Create a new TaskPool, with all tasks in non-spawned state.
pub const fn new() -> Self {
Self {
pool: [TaskStorage::NEW; N],
}
}
/// Try to spawn a task in the pool.
///
/// See [`TaskStorage::spawn()`] for details.
///
/// This will loop over the pool and spawn the task in the first storage that
/// is currently free. If none is free, a "poisoned" SpawnToken is returned,
/// which will cause [`Spawner::spawn()`](super::Spawner::spawn) to return the error.
pub fn spawn(&'static self, future: impl FnOnce() -> F) -> SpawnToken<impl Sized> {
for task in &self.pool {
if task.spawn_mark_used() {
return unsafe { SpawnToken::<F>::new(task.spawn_initialize(future)) };
}
}
SpawnToken::<F>::new_failed()
}
/// Like spawn(), but allows the task to be send-spawned if the args are Send even if
/// the future is !Send.
///
/// Not covered by semver guarantees. DO NOT call this directly. Intended to be used
/// by the Embassy macros ONLY.
///
/// SAFETY: `future` must be a closure of the form `move || my_async_fn(args)`, where `my_async_fn`
/// is an `async fn`, NOT a hand-written `Future`.
#[doc(hidden)]
pub unsafe fn _spawn_async_fn<FutFn>(&'static self, future: FutFn) -> SpawnToken<impl Sized>
where
FutFn: FnOnce() -> F,
{
// When send-spawning a task, we construct the future in this thread, and effectively
// "send" it to the executor thread by enqueuing it in its queue. Therefore, in theory,
// send-spawning should require the future `F` to be `Send`.
//
// The problem is this is more restrictive than needed. Once the future is executing,
// it is never sent to another thread. It is only sent when spawning. It should be
// enough for the task's arguments to be Send. (and in practice it's super easy to
// accidentally make your futures !Send, for example by holding an `Rc` or a `&RefCell` across an `.await`.)
//
// We can do it by sending the task args and constructing the future in the executor thread
// on first poll. However, this cannot be done in-place, so it'll waste stack space for a copy
// of the args.
//
// Luckily, an `async fn` future contains just the args when freshly constructed. So, if the
// args are Send, it's OK to send a !Send future, as long as we do it before first polling it.
//
// (Note: this is how the generators are implemented today, it's not officially guaranteed yet,
// but it's possible it'll be guaranteed in the future. See zulip thread:
// https://rust-lang.zulipchat.com/#narrow/stream/187312-wg-async/topic/.22only.20before.20poll.22.20Send.20futures )
//
// The `FutFn` captures all the args, so if it's Send, the task can be send-spawned.
// This is why we return `SpawnToken<FutFn>` below.
//
// This ONLY holds for `async fn` futures. The other `spawn` methods can be called directly
// by the user, with arbitrary hand-implemented futures. This is why these return `SpawnToken<F>`.
for task in &self.pool {
if task.spawn_mark_used() {
return SpawnToken::<FutFn>::new(task.spawn_initialize(future));
}
}
SpawnToken::<FutFn>::new_failed()
}
}
/// Raw executor.
///
/// This is the core of the Embassy executor. It is low-level, requiring manual
/// handling of wakeups and task polling. If you can, prefer using one of the
/// higher level executors in [`crate::executor`].
///
/// The raw executor leaves it up to you to handle wakeups and scheduling:
///
/// - To get the executor to do work, call `poll()`. This will poll all queued tasks (all tasks
/// that "want to run").
/// - You must supply a `signal_fn`. The executor will call it to notify you it has work
/// to do. You must arrange for `poll()` to be called as soon as possible.
///
/// `signal_fn` can be called from *any* context: any thread, any interrupt priority
/// level, etc. It may be called synchronously from any `Executor` method call as well.
/// You must deal with this correctly.
///
/// In particular, you must NOT call `poll` directly from `signal_fn`, as this violates
/// the requirement for `poll` to not be called reentrantly.
pub struct Executor {
run_queue: RunQueue,
signal_fn: fn(*mut ()),
signal_ctx: *mut (),
#[cfg(feature = "integrated-timers")]
pub(crate) timer_queue: timer_queue::TimerQueue,
#[cfg(feature = "integrated-timers")]
alarm: AlarmHandle,
}
impl Executor {
/// Create a new executor.
///
/// When the executor has work to do, it will call `signal_fn` with
/// `signal_ctx` as argument.
///
/// See [`Executor`] docs for details on `signal_fn`.
pub fn new(signal_fn: fn(*mut ()), signal_ctx: *mut ()) -> Self {
#[cfg(feature = "integrated-timers")]
let alarm = unsafe { unwrap!(driver::allocate_alarm()) };
#[cfg(feature = "integrated-timers")]
driver::set_alarm_callback(alarm, signal_fn, signal_ctx);
Self {
run_queue: RunQueue::new(),
signal_fn,
signal_ctx,
#[cfg(feature = "integrated-timers")]
timer_queue: timer_queue::TimerQueue::new(),
#[cfg(feature = "integrated-timers")]
alarm,
}
}
/// Enqueue a task in the task queue
///
/// # Safety
/// - `task` must be a valid pointer to a spawned task.
/// - `task` must be set up to run in this executor.
/// - `task` must NOT be already enqueued (in this executor or another one).
#[inline(always)]
unsafe fn enqueue(&self, cs: CriticalSection, task: NonNull<TaskHeader>) {
if self.run_queue.enqueue(cs, task) {
(self.signal_fn)(self.signal_ctx)
}
}
/// Spawn a task in this executor.
///
/// # Safety
///
/// `task` must be a valid pointer to an initialized but not-already-spawned task.
///
/// It is OK to use `unsafe` to call this from a thread that's not the executor thread.
/// In this case, the task's Future must be Send. This is because this is effectively
/// sending the task to the executor thread.
pub(super) unsafe fn spawn(&'static self, task: NonNull<TaskHeader>) {
task.as_ref().executor.set(self);
critical_section::with(|cs| {
self.enqueue(cs, task);
})
}
/// Poll all queued tasks in this executor.
///
/// This loops over all tasks that are queued to be polled (i.e. they're
/// freshly spawned or they've been woken). Other tasks are not polled.
///
/// You must call `poll` after receiving a call to `signal_fn`. It is OK
/// to call `poll` even when not requested by `signal_fn`, but it wastes
/// energy.
///
/// # Safety
///
/// You must NOT call `poll` reentrantly on the same executor.
///
/// In particular, note that `poll` may call `signal_fn` synchronously. Therefore, you
/// must NOT directly call `poll()` from your `signal_fn`. Instead, `signal_fn` has to
/// somehow schedule for `poll()` to be called later, at a time you know for sure there's
/// no `poll()` already running.
pub unsafe fn poll(&'static self) {
#[cfg(feature = "integrated-timers")]
self.timer_queue.dequeue_expired(Instant::now(), |task| wake_task(task));
self.run_queue.dequeue_all(|p| {
let task = p.as_ref();
#[cfg(feature = "integrated-timers")]
task.expires_at.set(Instant::MAX);
let state = task.state.fetch_and(!STATE_RUN_QUEUED, Ordering::AcqRel);
if state & STATE_SPAWNED == 0 {
// If task is not running, ignore it. This can happen in the following scenario:
// - Task gets dequeued, poll starts
// - While task is being polled, it gets woken. It gets placed in the queue.
// - Task poll finishes, returning done=true
// - RUNNING bit is cleared, but the task is already in the queue.
return;
}
// Run the task
task.poll_fn.read()(p as _);
// Enqueue or update into timer_queue
#[cfg(feature = "integrated-timers")]
self.timer_queue.update(p);
});
#[cfg(feature = "integrated-timers")]
{
// If this is already in the past, set_alarm will immediately trigger the alarm.
// This will cause `signal_fn` to be called, which will cause `poll()` to be called again,
// so we immediately do another poll loop iteration.
let next_expiration = self.timer_queue.next_expiration();
driver::set_alarm(self.alarm, next_expiration.as_ticks());
}
}
/// Get a spawner that spawns tasks in this executor.
///
/// It is OK to call this method multiple times to obtain multiple
/// `Spawner`s. You may also copy `Spawner`s.
pub fn spawner(&'static self) -> super::Spawner {
super::Spawner::new(self)
}
}
/// Wake a task by raw pointer.
///
/// You can obtain task pointers from `Waker`s using [`task_from_waker`].
///
/// # Safety
///
/// `task` must be a valid task pointer obtained from [`task_from_waker`].
pub unsafe fn wake_task(task: NonNull<TaskHeader>) {
critical_section::with(|cs| {
let header = task.as_ref();
let state = header.state.load(Ordering::Relaxed);
// If already scheduled, or if not started,
if (state & STATE_RUN_QUEUED != 0) || (state & STATE_SPAWNED == 0) {
return;
}
// Mark it as scheduled
header.state.store(state | STATE_RUN_QUEUED, Ordering::Relaxed);
// We have just marked the task as scheduled, so enqueue it.
let executor = &*header.executor.get();
executor.enqueue(cs, task);
})
}
#[cfg(feature = "integrated-timers")]
#[no_mangle]
unsafe fn _embassy_time_schedule_wake(at: Instant, waker: &core::task::Waker) {
let task = waker::task_from_waker(waker);
let task = task.as_ref();
let expires_at = task.expires_at.get();
task.expires_at.set(expires_at.min(at));
}

74
embassy-executor/src/raw/run_queue.rs Normal file
View File

@ -0,0 +1,74 @@
use core::ptr;
use core::ptr::NonNull;
use atomic_polyfill::{AtomicPtr, Ordering};
use critical_section::CriticalSection;
use super::TaskHeader;
pub(crate) struct RunQueueItem {
next: AtomicPtr<TaskHeader>,
}
impl RunQueueItem {
pub const fn new() -> Self {
Self {
next: AtomicPtr::new(ptr::null_mut()),
}
}
}
/// Atomic task queue using a very, very simple lock-free linked-list queue:
///
/// To enqueue a task, task.next is set to the old head, and head is atomically set to task.
///
/// Dequeuing is done in batches: the queue is emptied by atomically replacing head with
/// null. Then the batch is iterated following the next pointers until null is reached.
///
/// Note that batches will be iterated in the reverse order as they were enqueued. This is OK
/// for our purposes: it can't create fairness problems since the next batch won't run until the
/// current batch is completely processed, so even if a task enqueues itself instantly (for example
/// by waking its own waker) can't prevent other tasks from running.
pub(crate) struct RunQueue {
head: AtomicPtr<TaskHeader>,
}
impl RunQueue {
pub const fn new() -> Self {
Self {
head: AtomicPtr::new(ptr::null_mut()),
}
}
/// Enqueues an item. Returns true if the queue was empty.
///
/// # Safety
///
/// `item` must NOT be already enqueued in any queue.
#[inline(always)]
pub(crate) unsafe fn enqueue(&self, _cs: CriticalSection, task: NonNull<TaskHeader>) -> bool {
let prev = self.head.load(Ordering::Relaxed);
task.as_ref().run_queue_item.next.store(prev, Ordering::Relaxed);
self.head.store(task.as_ptr(), Ordering::Relaxed);
prev.is_null()
}
/// Empty the queue, then call `on_task` for each task that was in the queue.
/// NOTE: It is OK for `on_task` to enqueue more tasks. In this case they're left in the queue
/// and will be processed by the *next* call to `dequeue_all`, *not* the current one.
pub(crate) fn dequeue_all(&self, on_task: impl Fn(NonNull<TaskHeader>)) {
// Atomically empty the queue.
let mut ptr = self.head.swap(ptr::null_mut(), Ordering::AcqRel);
// Iterate the linked list of tasks that were previously in the queue.
while let Some(task) = NonNull::new(ptr) {
// If the task re-enqueues itself, the `next` pointer will get overwritten.
// Therefore, first read the next pointer, and only then process the task.
let next = unsafe { task.as_ref() }.run_queue_item.next.load(Ordering::Relaxed);
on_task(task);
ptr = next
}
}
}

85
embassy-executor/src/raw/timer_queue.rs Normal file
View File

@ -0,0 +1,85 @@
use core::cell::Cell;
use core::cmp::min;
use core::ptr;
use core::ptr::NonNull;
use atomic_polyfill::Ordering;
use embassy_time::Instant;
use super::{TaskHeader, STATE_TIMER_QUEUED};
pub(crate) struct TimerQueueItem {
next: Cell<*mut TaskHeader>,
}
impl TimerQueueItem {
pub const fn new() -> Self {
Self {
next: Cell::new(ptr::null_mut()),
}
}
}
pub(crate) struct TimerQueue {
head: Cell<*mut TaskHeader>,
}
impl TimerQueue {
pub const fn new() -> Self {
Self {
head: Cell::new(ptr::null_mut()),
}
}
pub(crate) unsafe fn update(&self, p: NonNull<TaskHeader>) {
let task = p.as_ref();
if task.expires_at.get() != Instant::MAX {
let old_state = task.state.fetch_or(STATE_TIMER_QUEUED, Ordering::AcqRel);
let is_new = old_state & STATE_TIMER_QUEUED == 0;
if is_new {
task.timer_queue_item.next.set(self.head.get());
self.head.set(p.as_ptr());
}
}
}
pub(crate) unsafe fn next_expiration(&self) -> Instant {
let mut res = Instant::MAX;
self.retain(|p| {
let task = p.as_ref();
let expires = task.expires_at.get();
res = min(res, expires);
expires != Instant::MAX
});
res
}
pub(crate) unsafe fn dequeue_expired(&self, now: Instant, on_task: impl Fn(NonNull<TaskHeader>)) {
self.retain(|p| {
let task = p.as_ref();
if task.expires_at.get() <= now {
on_task(p);
false
} else {
true
}
});
}
pub(crate) unsafe fn retain(&self, mut f: impl FnMut(NonNull<TaskHeader>) -> bool) {
let mut prev = &self.head;
while !prev.get().is_null() {
let p = NonNull::new_unchecked(prev.get());
let task = &*p.as_ptr();
if f(p) {
// Skip to next
prev = &task.timer_queue_item.next;
} else {
// Remove it
prev.set(task.timer_queue_item.next.get());
task.state.fetch_and(!STATE_TIMER_QUEUED, Ordering::AcqRel);
}
}
}
}

33
embassy-executor/src/raw/util.rs Normal file
View File

@ -0,0 +1,33 @@
use core::cell::UnsafeCell;
use core::mem::MaybeUninit;
use core::ptr;
pub(crate) struct UninitCell<T>(MaybeUninit<UnsafeCell<T>>);
impl<T> UninitCell<T> {
pub const fn uninit() -> Self {
Self(MaybeUninit::uninit())
}
pub unsafe fn as_mut_ptr(&self) -> *mut T {
(*self.0.as_ptr()).get()
}
#[allow(clippy::mut_from_ref)]
pub unsafe fn as_mut(&self) -> &mut T {
&mut *self.as_mut_ptr()
}
pub unsafe fn write(&self, val: T) {
ptr::write(self.as_mut_ptr(), val)
}
pub unsafe fn drop_in_place(&self) {
ptr::drop_in_place(self.as_mut_ptr())
}
}
impl<T: Copy> UninitCell<T> {
pub unsafe fn read(&self) -> T {
ptr::read(self.as_mut_ptr())
}
}

53
embassy-executor/src/raw/waker.rs Normal file
View File

@ -0,0 +1,53 @@
use core::mem;
use core::ptr::NonNull;
use core::task::{RawWaker, RawWakerVTable, Waker};
use super::{wake_task, TaskHeader};
const VTABLE: RawWakerVTable = RawWakerVTable::new(clone, wake, wake, drop);
unsafe fn clone(p: *const ()) -> RawWaker {
RawWaker::new(p, &VTABLE)
}
unsafe fn wake(p: *const ()) {
wake_task(NonNull::new_unchecked(p as *mut TaskHeader))
}
unsafe fn drop(_: *const ()) {
// nop
}
pub(crate) unsafe fn from_task(p: NonNull<TaskHeader>) -> Waker {
Waker::from_raw(RawWaker::new(p.as_ptr() as _, &VTABLE))
}
/// Get a task pointer from a waker.
///
/// This can be used as an optimization in wait queues to store task pointers
/// (1 word) instead of full Wakers (2 words). This saves a bit of RAM and helps
/// avoid dynamic dispatch.
///
/// You can use the returned task pointer to wake the task with [`wake_task`](super::wake_task).
///
/// # Panics
///
/// Panics if the waker is not created by the Embassy executor.
pub fn task_from_waker(waker: &Waker) -> NonNull<TaskHeader> {
// safety: OK because WakerHack has the same layout as Waker.
// This is not really guaranteed because the structs are `repr(Rust)`, it is
// indeed the case in the current implementation.
// TODO use waker_getters when stable. https://github.com/rust-lang/rust/issues/96992
let hack: &WakerHack = unsafe { mem::transmute(waker) };
if hack.vtable != &VTABLE {
panic!("Found waker not created by the Embassy executor. `embassy_time::Timer` only works with the Embassy executor.")
}
// safety: we never create a waker with a null data pointer.
unsafe { NonNull::new_unchecked(hack.data as *mut TaskHeader) }
}
struct WakerHack {
data: *const (),
vtable: &'static RawWakerVTable,
}