dtact 0.2.5

Dtact: A non-preemptive, stackful coroutine runtime featuring a lock-free context arena, P2P mesh scheduling, and architecture-specific assembly switchers. Designed for hardware-level control and non-blocking heterogeneous orchestration.
Documentation
#![allow(dead_code)]
use dtact::{DtactWaitExt, Priority, WorkloadKind, spawn, spawn_with, yield_now};
use std::sync::Arc;
use std::sync::atomic::{AtomicU32, Ordering};

#[dtact::dtact_init(workers = 4, capacity = 2048, safety = "Safety1")]
#[cfg_attr(miri, ignore)]
#[test]
fn test_dtact_comprehensive_e2e() {
    // 1. Test Fiber Execution
    {
        let counter = Arc::new(AtomicU32::new(0));
        let c = counter.clone();
        spawn(async move {
            c.fetch_add(1, Ordering::SeqCst);
            yield_now().await;
            c.fetch_add(1, Ordering::SeqCst);
        });
        std::thread::sleep(std::time::Duration::from_millis(100));
        assert_eq!(counter.load(Ordering::SeqCst), 2);
    }

    // 2. Test Workload Kind and Priority
    {
        let counter = Arc::new(AtomicU32::new(0));
        let c_high = counter.clone();
        spawn_with()
            .priority(Priority::High)
            .kind(WorkloadKind::Compute)
            .name("high-priority-compute")
            .spawn(async move {
                c_high.fetch_add(10, Ordering::SeqCst);
            });

        let c_low = counter.clone();
        spawn_with()
            .priority(Priority::Low)
            .kind(WorkloadKind::IO)
            .name("low-priority-io")
            .spawn(async move {
                c_low.fetch_add(1, Ordering::SeqCst);
            });

        std::thread::sleep(std::time::Duration::from_millis(100));
        assert!(counter.load(Ordering::SeqCst) >= 11);
    }

    // 3. Test Zero-Copy Fallback
    {
        struct LargeFuture([u8; 16384]);
        impl core::future::Future for LargeFuture {
            type Output = ();
            fn poll(
                self: core::pin::Pin<&mut Self>,
                _: &mut core::task::Context<'_>,
            ) -> core::task::Poll<Self::Output> {
                core::task::Poll::Ready(())
            }
        }

        let initial_escaped = dtact::HEAP_ESCAPED_SPAWNS.load(Ordering::Relaxed);
        spawn(LargeFuture([0; 16384]));
        std::thread::sleep(std::time::Duration::from_millis(100));
        let final_escaped = dtact::HEAP_ESCAPED_SPAWNS.load(Ordering::Relaxed);
        assert!(
            final_escaped > initial_escaped,
            "Large future should have escaped to heap"
        );
    }

    // 4. Test DtactWaitExt Bridge
    {
        let counter = Arc::new(AtomicU32::new(0));
        let c = counter.clone();
        async fn some_async_val() -> u32 {
            100
        }
        spawn(async move {
            let val = some_async_val().wait();
            c.store(val, Ordering::SeqCst);
        });
        std::thread::sleep(std::time::Duration::from_millis(100));
        assert_eq!(counter.load(Ordering::SeqCst), 100);
    }

    // 5. Test #[task] macro execution and metadata injection
    {
        let res1 = Arc::new(AtomicU32::new(0));
        let res2 = Arc::new(std::sync::Mutex::new(String::new()));

        let handle1 = spawn!(custom_task_1(40, 2, res1.clone()));
        let handle2 = spawn!(custom_task_2(res2.clone()));

        std::thread::sleep(std::time::Duration::from_millis(100));
        dtact::dtact_await(handle1);
        dtact::dtact_await(handle2);

        assert_eq!(res1.load(Ordering::SeqCst), 42);
        assert_eq!(*res2.lock().unwrap(), "hello_from_task");
    }
}

#[dtact::task(priority = "High", kind = "Compute", switcher = "CrossThreadNoFloat")]
async fn custom_task_1(x: u32, y: u32, output: Arc<AtomicU32>) {
    output.store(x + y, Ordering::SeqCst);
}

#[dtact::task(
    priority = "Low",
    kind = "IO",
    switcher = "SameThreadFloat",
    affinity = "SameCore"
)]
async fn custom_task_2(output: Arc<std::sync::Mutex<String>>) {
    let mut lock = output.lock().unwrap();
    *lock = "hello_from_task".to_string();
}