Trait scale_info::prelude::marker::Send

1.0.0 · source · [−]

pub unsafe auto trait Send { }

Expand description

Types that can be transferred across thread boundaries.

This trait is automatically implemented when the compiler determines it’s appropriate.

An example of a non-Send type is the reference-counting pointer rc::Rc. If two threads attempt to clone Rcs that point to the same reference-counted value, they might try to update the reference count at the same time, which is undefined behavior because Rc doesn’t use atomic operations. Its cousin sync::Arc does use atomic operations (incurring some overhead) and thus is Send.

See the Nomicon for more details.

Implementations on Foreign Types

impl Send for Argument

impl Send for FormatSpec

impl Send for Alignment

impl Send for Count

source

impl<T> Send for Receiver<T> where
T: Send,

1.44.0 · source

impl<'a> Send for IoSliceMut<'a>

source

impl<T> Send for Sender<T> where
T: Send,

1.26.0 · source

impl<'_, T> !Send for MutexGuard<'_, T> where
T: ?Sized,

1.44.0 · source

impl<'a> Send for IoSlice<'a>

source

impl Send for Once

1.29.0 · source

impl<T> Send for JoinHandle<T>

source

impl<T> Send for SyncOnceCell<T> where
T: Send,

1.26.0 · source

impl !Send for ArgsOs

source

impl<T> !Send for *const T where
T: ?Sized,

source

impl<T> !Send for *mut T where
T: ?Sized,

1.25.0 · source

impl<T> !Send for NonNull<T> where
T: ?Sized,

NonNull pointers are not Send because the data they reference may be aliased.

source

impl<T> Send for AtomicPtr<T>

source

impl<'_, T> Send for Iter<'_, T> where
T: Sync,

source

impl<'_, T> Send for IterMut<'_, T> where
T: Send,

source

impl<T> Send for RefCell<T> where
T: Send + ?Sized,

source

impl<'_, T> Send for &'_ T where
T: Sync + ?Sized,

source

impl<'_, T> Send for &'_ mut T where
T: Send + ?Sized,

source

impl<T> Send for Cell<T> where
T: Send + ?Sized,

1.36.0 · source

impl Send for Waker

source

impl<Dyn> Send for DynMetadata<Dyn> where
Dyn: ?Sized,

1.4.0 · source

impl<T> Send for Weak<T> where
T: Sync + Send + ?Sized,

1.4.0 · source

impl<T> !Send for Weak<T> where
T: ?Sized,

source

impl<T> !Send for Rc<T> where
T: ?Sized,

source

impl<T> Send for Arc<T> where
T: Sync + Send + ?Sized,

source

impl<'a, T, const CAP: usize> Send for Drain<'a, T, CAP> where
T: Send,

source

impl<'_, O, T> Send for Drain<'_, O, T> where
O: BitOrder,
T: BitStore,

source

impl<'_, O, T> Send for Iter<'_, O, T> where
O: BitOrder,
T: BitStore,

source

impl<O, T> Send for BitSlice<O, T> where
O: BitOrder,
T: BitStore + Sync,

Conditionally mark BitSlice as Send based on its T type argument.

In order for BitSlice to be Send (that is, &mut BitSlice can be moved across thread boundaries), it must be capable of writing to memory without invalidating any other &BitSlice handles that alias the same memory address.

This is true when T is one of the fundamental integers, because no other &BitSlice handle is able to observe mutations, or when T is a BitSafe type that implements atomic read-modify-write instructions, because other &BitSlice types will be protected from data races by the hardware.

When T is a non-atomic BitSafe type, BitSlice cannot be Send, because one &mut BitSlice moved across a thread boundary may cause mutation that another &BitSlice may observe, but the instructions used to access memory do not guard against data races.

A &mut BitSlice over aliased memory addresses is equivalent to either a &Cell or &AtomicT, depending on what the radium crate makes available for the register width.

source