Getting deadlock inside match of async function - asynchronous

I'm getting a deadlock on the following example:
use tokio::net::TcpListener;
use tokio::io::{AsyncReadExt, AsyncWriteExt};
use futures::lock::Mutex;
use std::sync::Arc;
struct A{
}
impl A {
pub async fn do_something(&self) -> std::result::Result<(), ()>{
Err(())
}
}
async fn lock_and_use(a: Arc<Mutex<A>>) {
match a.clone().lock().await.do_something().await {
Ok(()) => {
},
Err(()) => {
//try again on error
println!("trying again");
a.clone().lock().await.do_something().await.unwrap();
}
}
}
#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
println!("begin");
let a = Arc::new(Mutex::new(A{}));
lock_and_use(a.clone()).await;
println!("end");
Ok(())
}
Playground
If I did this:
a.clone().lock().await.do_something().await;
a.clone().lock().await.do_something().await;
there would be no problem, the lock() dies on the same line it's created. I thought this principle would be the same for match. If you think about match, it locks the value, calls do_something, awaits on it, and then compares the value. It's true that do_something returns a future which capture the self, but when we await on it, it should discard the self. Why it still holds self? How can I solve this without cloning the result?

Yes:
Temporaries live for the entire statement, never shorter.
Cause code could be:
{
match self.cache.read() { // <-- direct pattern matching
Ok(ref data) => Some(data)
_ => None,
}
}.map(|data| {
// use `data` - the lock better be held
})
Read this issue for more detail.
So you need to lock outside your match statement:
let x = a.clone().lock().await.do_something().await;
match x {
Ok(()) => {}
Err(()) => {
a.clone().lock().await.do_something().await.unwrap();
}
}

Related

Rust - return a future in a closure

I am trying to add a return type of a future in a closure. But the compiler is telling me that
`impl Trait` only allowed in function and inherent method return types, not in closure return
I have have also tried wrapping it in Box but that didn't work. I tried type aliases but they are nighly only features. I am unable to understand how else can I solve it.
pub async fn execute_event_handler(
event_handler: Option<Arc<PyFunction>>,
event_loop: Arc<Py<PyAny>>,
) -> Result<(), Box<dyn std::error::Error>> {
if let Some(handler) = event_handler {
match &(*handler) {
PyFunction::SyncFunction(function) => {
println!("Startup event handler");
Python::with_gil(|py| -> Result<(), Box<dyn std::error::Error>> {
function.call0(py)?;
Ok(())
})?;
}
PyFunction::CoRoutine(function) => {
let future = Python::with_gil(
|py| -> impl Future<Output = Result<Py<PyAny>, PyErr>> + Send { // this is giving an error
println!("Startup event handler async");
let coroutine = function.as_ref(py).call0().unwrap();
pyo3_asyncio::into_future_with_loop((*event_loop).as_ref(py), coroutine)
.unwrap()
},
);
future.await?;
}
}
}
Ok(())
}
PS: I want to wrap the closure output in an Result type and hence a return type is necessary.
Don't annotate the future.
If you need to annotate the enclosing type (e.g. Result), you can either annotate it when returning it (Result::<_, ErrorType>) or as the return type, but let inference find the future type itself with _: || -> Result<_, ErrorType>.

How to run multiple Tokio async tasks in a loop without using tokio::spawn?

I built a LED clock that also displays weather. My program does a couple of different things in a loop, each thing with a different interval:
updates the LEDs every 50ms,
checks the light level (to adjust the brightness) every 1 second,
fetches weather every 10 minutes,
actually some more, but that's irrelevant.
Updating the LEDs is the most critical: I don't want this to be delayed when e.g. weather is being fetched. This should not be a problem as fetching weather is mostly an async HTTP call.
Here's the code that I have:
let mut measure_light_stream = tokio::time::interval(Duration::from_secs(1));
let mut update_weather_stream = tokio::time::interval(WEATHER_FETCH_INTERVAL);
let mut update_leds_stream = tokio::time::interval(UPDATE_LEDS_INTERVAL);
loop {
tokio::select! {
_ = measure_light_stream.tick() => {
let light = lm.get_light();
light_smooth.sp = light;
},
_ = update_weather_stream.tick() => {
let fetched_weather = weather_service.get(&config).await;
// Store the fetched weather for later access from the displaying function.
weather_clock.weather = fetched_weather.clone();
},
_ = update_leds_stream.tick() => {
// Some code here that actually sets the LEDs.
// This code accesses the weather_clock, the light level etc.
},
}
}
I realised the code doesn't do what I wanted it to do - fetching the weather blocks the execution of the loop. I see why - the docs of tokio::select! say the other branches are cancelled as soon as the update_weather_stream.tick() expression completes.
How do I do this in such a way that while fetching the weather is waiting on network, the LEDs are still updated? I figured out I could use tokio::spawn to start a separate non-blocking "thread" for fetching weather, but then I have problems with weather_service not being Send, let alone weather_clock not being shareable between threads. I don't want this complication, I'm fine with everything running in a single thread, just like what select! does.
Reproducible example
use std::time::Duration;
use tokio::time::{interval, sleep};
#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
let mut slow_stream = interval(Duration::from_secs(3));
let mut fast_stream = interval(Duration::from_millis(200));
// Note how access to this data is straightforward, I do not want
// this to get more complicated, e.g. care about threads and Send.
let mut val = 1;
loop {
tokio::select! {
_ = fast_stream.tick() => {
println!(".{}", val);
},
_ = slow_stream.tick() => {
println!("Starting slow operation...");
// The problem: During this await the dots are not printed.
sleep(Duration::from_secs(1)).await;
val += 1;
println!("...done");
},
}
}
}
You can use tokio::join! to run multiple async operations concurrently within the same task.
Here's an example:
async fn measure_light(halt: &Cell<bool>) {
while !halt.get() {
let light = lm.get_light();
// ....
tokio::time::sleep(Duration::from_secs(1)).await;
}
}
async fn blink_led(halt: &Cell<bool>) {
while !halt.get() {
// LED blinking code
tokio::time::sleep(UPDATE_LEDS_INTERVAL).await;
}
}
async fn poll_weather(halt: &Cell<bool>) {
while !halt.get() {
let weather = weather_service.get(&config).await;
// ...
tokio::time::sleep(WEATHER_FETCH_INTERVAL).await;
}
}
// example on how to terminate execution
async fn terminate(halt: &Cell<bool>) {
tokio::time::sleep(Duration::from_secs(10)).await;
halt.set(true);
}
async fn main() {
let halt = Cell::new(false);
tokio::join!(
measure_light(&halt),
blink_led(&halt),
poll_weather(&halt),
terminate(&halt),
);
}
If you're using tokio::TcpStream or other non-blocking IO, then it should allow for concurrent execution.
I've added a Cell flag for halting execution as an example. You can use the same technique to share any mutable state between join branches.
EDIT: Same thing can be done with tokio::select!. The main difference with your code is that the actual "business logic" is inside the futures awaited by select.
select allows you to drop unfinished futures instead of waiting for them to exit on their own (so halt termination flag is not necessary).
async fn main() {
tokio::select! {
_ = measure_light() => {},
_ = blink_led() = {},
_ = poll_weather() => {},
}
}
Here's a concrete solution, based on the second part of stepan's answer:
use std::time::Duration;
use tokio::time::sleep;
#[tokio::main]
async fn main() {
// Cell is an acceptable complication when accessing the data.
let val = std::cell::Cell::new(1);
tokio::select! {
_ = async {loop {
println!(".{}", val.get());
sleep(Duration::from_millis(200)).await;
}} => {},
_ = async {loop {
println!("Starting slow operation...");
// The problem: During this await the dots are not printed.
sleep(Duration::from_secs(1)).await;
val.set(val.get() + 1);
println!("...done");
sleep(Duration::from_secs(3)).await;
}} => {},
}
}
Playground link

Not noticing any performance improvement when using async

I have a small program that executes the aws s3 cli commands but with different arguments. I'm using the Command crate and the the command makes a network call and returns some response. At first I have this synchronous & single-threaded implementation:
fn make_call<'a>(_name: &'a str, _bucket_poll: &mut BucketPoll<'a>) -> Option<BucketDetails<'a>> {
let invoke_result = invoke_network_call(_name);
let mut bucket = BucketDetails::new(_name);
match invoke_result {
Ok(invoke_str) => {
bucket.output = invoke_str;
_bucket_poll.insert_bucket(bucket.clone());
_bucket_poll.successful_count += 1;
Some(bucket)
}
Err(_) => {
_bucket_poll.insert_bucket(bucket);
None
}
}
}
// I invoke this function in sequential order, something like
make_call('name_1');
make_call('name_2');
make_call('name_3');
Because I don't really care at which order this function is executed, I decided to learn Tokio to help with performance. I changed the make_call function to be async:
async fn make_call_race() -> ExecutionResult {
let bucket_poll = BucketPoll::new();
let bucket_poll_guard = Arc::new(Mutex::new(bucket_poll));
loop {
let bucket_details = tokio::select! {
Some(bucket_details) = make_call_async("name_1", &bucket_poll_guard) => bucket_details,
Some(bucket_details) = make_call_async("name_2", &bucket_poll_guard) => bucket_details,
Some(bucket_details) = make_call_async("name_3", &bucket_poll_guard) => bucket_details,
Some(bucket_details) = make_call_async("name_4", &bucket_poll_guard) => bucket_details,
else => { break }
};
success_printer(bucket_details);
}
// more printing, no more network calls
ExecutionResult::Success
}
make_call_async is essentially the same as make_call:
async fn make_call_async<'a>(
_name: &'a str,
_bucket_poll_guard: &'a Arc<Mutex<BucketPoll<'a>>>,
) -> Option<BucketDetails<'a>> {
{
if let Ok(bucket_poll_guard) = _bucket_poll_guard.lock() {
if bucket_poll_guard.has_polled(_name) {
return None;
}
}
}
let invoke_result = invoke_network_call(_name);
let mut bucket = BucketDetails::new(_name);
match invoke_result {
Ok(invoke_str) => {
bucket.output = invoke_str;
{
if let Ok(mut bucket_poll_guard) = _bucket_poll_guard.lock() {
bucket_poll_guard.insert_bucket(bucket.clone());
bucket_poll_guard.successful_count += 1;
}
}
Some(bucket)
}
Err(_) => {
{
if let Ok(mut bucket_poll_guard) = _bucket_poll_guard.lock() {
bucket_poll_guard.insert_bucket(bucket);
}
}
None
}
}
}
When I run the async version, I do see that my network calls are made a random order but I do not notice any speedups. I increased the number of network calls to ~50ish invocations but the runtime is nearly the same if not slightly worse. As I am new to async programming and Rust in general, I would like to understand why my async implementation does not seem to offer any improvement.
Extra:
Here is the invoke_network_call method:
fn invoke_network_call(_name: &str) -> core::result::Result<String, AwsCliError> {
let output = Command::new("aws")
.arg("s3")
.arg("ls")
.arg(_name)
.output()
.expect("Could not list s3 objects");
if !output.status.success() {
err_printer(format!("Failed to list s3 objects for bucket {}.", _name));
return Err(AwsCliError);
}
let output_str = get_stdout_string_from_output(&output);
Ok(output_str)
}
EDIT: yorodm's comment makes sense. What I did was use Tokio's Command instead of std::process's Command and made the invoke_network_call async. This reduced my runtime by half. Thank you!
You could rewrite invoke_network_call using an async version of Command.
async fn invoke_network_call(_name: &str) -> core::result::Result<String, AwsCliError> {
let output = tokio::process::Command::new("aws")
.arg("s3")
.arg("ls")
.arg(_name)
.output()
.await
.expect("Could not list s3 objects");
if !output.status.success() {
err_printer(format!("Failed to list s3 objects for bucket {}.", _name));
return Err(AwsCliError);
}
let output_str = get_stdout_string_from_output(&output);
Ok(output_str)
}
Thus removing the blocking std::process::Command call. However I would say that if you're going to access AWS services you should go with rusoto

Is there a way to do recursive async calls on Rust without Box?

Recursive calls with a retry count embedded into the argument is good for the mental model of the project because you don't have to keep thinking about the state of the object. The retry_count is passed in every call. Here's a simple implementation:
use std::future::Future;
use futures::future::{BoxFuture, FutureExt};
fn do_or_fail() -> std::result::Result<(),()> {
Ok(())
}
fn do_something<'a>(
retry_count: u32
) -> BoxFuture<'a, std::result::Result<(), ()>> {
async move {
match do_or_fail() {
Ok(_) => Ok(()),
Err(_) => do_something(retry_count -1).await
}
}.boxed()
}
fn main() {
do_something(3);
}
Playground
The problem is that this requires a dynamic allocation at every call so it can return the BoxFuture. This is because of how async is implemented in Rust. It generates a state machine for every await call so the return type is not Sized.
What would be good ways to overcome dynamic allocation in recursive async calls?
While it technically doesn't overcome the problem (it just hides it), I can recommend the crate async_recursion.
https://crates.io/crates/async-recursion
use async_recursion::async_recursion;
use tokio;
fn do_or_fail() -> std::result::Result<(), ()> {
Ok(())
}
#[async_recursion]
async fn do_something(retry_count: u32) -> std::result::Result<(), ()> {
match do_or_fail() {
Ok(_) => Ok(()),
Err(_) => do_something(retry_count - 1).await,
}
}
#[tokio::main]
async fn main() {
do_something(3).await.unwrap()
}
I don't think you actually can overcome dynamic allocation in async calls, that's just how async works in general.

Why does tokio::spawn have a delay when called next to crossbeam_channel::select?

I'm creating a task which will spawn other tasks. Some of them will take some time, so they cannot be awaited, but they can run in parallel:
src/main.rs
use crossbeam::crossbeam_channel::{bounded, select};
#[tokio::main]
async fn main() {
let (s, r) = bounded::<usize>(1);
tokio::spawn(async move {
let mut counter = 0;
loop {
let loop_id = counter.clone();
tokio::spawn(async move { // why this one was not fired?
println!("inner task {}", loop_id);
}); // .await.unwrap(); - solves issue, but this is long task which cannot be awaited
println!("loop {}", loop_id);
select! {
recv(r) -> rr => {
// match rr {
// Ok(ee) => {
// println!("received from channel {}", loop_id);
// tokio::spawn(async move {
// println!("received from channel task {}", loop_id);
// });
// },
// Err(e) => println!("{}", e),
// };
},
// more recv(some_channel) ->
}
counter = counter + 1;
}
});
// let s_clone = s.clone();
// tokio::spawn(async move {
// s_clone.send(2).unwrap();
// });
loop {
// rest of the program
}
}
I've noticed strange behavior. This outputs:
loop 0
I was expecting it to also output inner task 0.
If I send a value to channel, the output will be:
loop 0
inner task 0
loop 1
This is missing inner task 1.
Why is inner task spawned with one loop of delay?
The first time I noticed such behavior with 'received from channel task' delayed one loop, but when I reduced code to prepare sample this started to happen with 'inner task'. It might be worth mentioning that if I write second tokio::spawn right to another, only the last one will have this issue. Is there something I should be aware when calling tokio::spawn and select!? What causes this one loop of delay?
Cargo.toml dependencies
[dependencies]
tokio = { version = "0.2", features = ["full"] }
crossbeam = "0.7"
Rust 1.46, Windows 10
select! is blocking, and the docs for tokio::spawn say:
The spawned task may execute on the current thread, or it may be sent to a different thread to be executed.
In this case, the select! "future" is actually a blocking function, and spawn doesn't use a new thread (either in the first invocation or the one inside the loop).
Because you don't tell tokio that you are going to block, tokio doesn't think another thread is needed (from tokio's perspective, you only have 3 futures which should never block, so why would you need another thread anyway?).
The solution is to use the tokio::task::spawn_blocking for the select!-ing closure (which will no longer be a future, so async move {} is now move || {}).
Now tokio will know that this function actually blocks, and will move it to another thread (while keeping all the actual futures in other execution threads).
use crossbeam::crossbeam_channel::{bounded, select};
#[tokio::main]
async fn main() {
let (s, r) = bounded::<usize>(1);
tokio::task::spawn_blocking(move || {
// ...
});
loop {
// rest of the program
}
}
Link to playground
Another possible solution is to use a non-blocking channel like tokio::sync::mpsc, on which you can use await and get the expected behavior, like this playground example with direct recv().await or with tokio::select!, like this:
use tokio::sync::mpsc;
#[tokio::main]
async fn main() {
let (mut s, mut r) = mpsc::channel::<usize>(1);
tokio::spawn(async move {
loop {
// ...
tokio::select! {
Some(i) = r.recv() => {
println!("got = {}", i);
}
}
}
});
loop {
// rest of the program
}
}
Link to playground

Resources