Struct Atomic 

pub struct Atomic<T: AtomicPrimitive> { /* private fields */ }

🔬This is a nightly-only experimental API. (generic_atomic #130539)

A memory location which can be safely modified from multiple threads.

This has the same size and bit validity as the underlying type T. However, the alignment of this type is always equal to its size, even on targets where T has alignment less than its size.

For more about the differences between atomic types and non-atomic types as well as information about the portability of this type, please see the module-level documentation.

Note: This type is only available on platforms that support atomic loads and stores of T.

Implementations

impl Atomic<bool>

pub const fn new(v: bool) -> AtomicBool

Creates a new AtomicBool.

Examples

use std::sync::atomic::AtomicBool;

let atomic_true = AtomicBool::new(true);
let atomic_false = AtomicBool::new(false);

pub const unsafe fn from_ptr<'a>(ptr: *mut bool) -> &'a AtomicBool

Creates a new AtomicBool from a pointer.

Examples

use std::sync::atomic::{self, AtomicBool};

// Get a pointer to an allocated value
let ptr: *mut bool = Box::into_raw(Box::new(false));

assert!(ptr.cast::<AtomicBool>().is_aligned());

{
    // Create an atomic view of the allocated value
    let atomic = unsafe { AtomicBool::from_ptr(ptr) };

    // Use `atomic` for atomic operations, possibly share it with other threads
    atomic.store(true, atomic::Ordering::Relaxed);
}

// It's ok to non-atomically access the value behind `ptr`,
// since the reference to the atomic ended its lifetime in the block above
assert_eq!(unsafe { *ptr }, true);

// Deallocate the value
unsafe { drop(Box::from_raw(ptr)) }

Safety

• ptr must be aligned to align_of::<AtomicBool>() (note that this is always true, since align_of::<AtomicBool>() == 1).
• ptr must be valid for both reads and writes for the whole lifetime 'a.
• You must adhere to the Memory model for atomic accesses. In particular, it is not allowed to mix conflicting atomic and non-atomic accesses, or atomic accesses of different sizes, without synchronization.

pub fn get_mut(&mut self) -> &mut bool

Returns a mutable reference to the underlying bool.

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::{AtomicBool, Ordering};

let mut some_bool = AtomicBool::new(true);
assert_eq!(*some_bool.get_mut(), true);
*some_bool.get_mut() = false;
assert_eq!(some_bool.load(Ordering::SeqCst), false);

pub fn from_mut(v: &mut bool) -> &mut Self

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Gets atomic access to a &mut bool.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicBool, Ordering};

let mut some_bool = true;
let a = AtomicBool::from_mut(&mut some_bool);
a.store(false, Ordering::Relaxed);
assert_eq!(some_bool, false);

pub fn get_mut_slice(this: &mut [Self]) -> &mut [bool]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Gets non-atomic access to a &mut [AtomicBool] slice.

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicBool, Ordering};

let mut some_bools = [const { AtomicBool::new(false) }; 10];

let view: &mut [bool] = AtomicBool::get_mut_slice(&mut some_bools);
assert_eq!(view, [false; 10]);
view[..5].copy_from_slice(&[true; 5]);

std::thread::scope(|s| {
    for t in &some_bools[..5] {
        s.spawn(move || assert_eq!(t.load(Ordering::Relaxed), true));
    }

    for f in &some_bools[5..] {
        s.spawn(move || assert_eq!(f.load(Ordering::Relaxed), false));
    }
});

pub fn from_mut_slice(v: &mut [bool]) -> &mut [Self]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Gets atomic access to a &mut [bool] slice.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicBool, Ordering};

let mut some_bools = [false; 10];
let a = &*AtomicBool::from_mut_slice(&mut some_bools);
std::thread::scope(|s| {
    for i in 0..a.len() {
        s.spawn(move || a[i].store(true, Ordering::Relaxed));
    }
});
assert_eq!(some_bools, [true; 10]);

pub const fn into_inner(self) -> bool

Consumes the atomic and returns the contained value.

This is safe because passing self by value guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::AtomicBool;

let some_bool = AtomicBool::new(true);
assert_eq!(some_bool.into_inner(), true);

pub fn load(&self, order: Ordering) -> bool

Loads a value from the bool.

load takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Acquire and Relaxed.

Panics

Panics if order is Release or AcqRel.

Examples

use std::sync::atomic::{AtomicBool, Ordering};

let some_bool = AtomicBool::new(true);

assert_eq!(some_bool.load(Ordering::Relaxed), true);

pub fn store(&self, val: bool, order: Ordering)

Stores a value into the bool.

store takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Release and Relaxed.

Panics

Panics if order is Acquire or AcqRel.

Examples

use std::sync::atomic::{AtomicBool, Ordering};

let some_bool = AtomicBool::new(true);

some_bool.store(false, Ordering::Relaxed);
assert_eq!(some_bool.load(Ordering::Relaxed), false);

pub fn swap(&self, val: bool, order: Ordering) -> bool

Stores a value into the bool, returning the previous value.

swap takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicBool, Ordering};

let some_bool = AtomicBool::new(true);

assert_eq!(some_bool.swap(false, Ordering::Relaxed), true);
assert_eq!(some_bool.load(Ordering::Relaxed), false);

pub fn compare_and_swap( &self, current: bool, new: bool, order: Ordering, ) -> bool

👎Deprecated since 1.50.0: Use compare_exchange or compare_exchange_weak instead

Stores a value into the bool if the current value is the same as the current value.

The return value is always the previous value. If it is equal to current, then the value was updated.

compare_and_swap also takes an Ordering argument which describes the memory ordering of this operation. Notice that even when using AcqRel, the operation might fail and hence just perform an Acquire load, but not have Release semantics. Using Acquire makes the store part of this operation Relaxed if it happens, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Migrating to compare_exchange and compare_exchange_weak

compare_and_swap is equivalent to compare_exchange with the following mapping for memory orderings:

Original	Success	Failure
Relaxed	Relaxed	Relaxed
Acquire	Acquire	Acquire
Release	Release	Relaxed
AcqRel	AcqRel	Acquire
SeqCst	SeqCst	SeqCst

compare_and_swap and compare_exchange also differ in their return type. You can use compare_exchange(...).unwrap_or_else(|x| x) to recover the behavior of compare_and_swap, but in most cases it is more idiomatic to check whether the return value is Ok or Err rather than to infer success vs failure based on the value that was read.

During migration, consider whether it makes sense to use compare_exchange_weak instead. compare_exchange_weak is allowed to fail spuriously even when the comparison succeeds, which allows the compiler to generate better assembly code when the compare and swap is used in a loop.

Examples

use std::sync::atomic::{AtomicBool, Ordering};

let some_bool = AtomicBool::new(true);

assert_eq!(some_bool.compare_and_swap(true, false, Ordering::Relaxed), true);
assert_eq!(some_bool.load(Ordering::Relaxed), false);

assert_eq!(some_bool.compare_and_swap(true, true, Ordering::Relaxed), false);
assert_eq!(some_bool.load(Ordering::Relaxed), false);

pub fn compare_exchange( &self, current: bool, new: bool, success: Ordering, failure: Ordering, ) -> Result<bool, bool>

Stores a value into the bool if the current value is the same as the current value.

The return value is a result indicating whether the new value was written and containing the previous value. On success this value is guaranteed to be equal to current.

compare_exchange takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicBool, Ordering};

let some_bool = AtomicBool::new(true);

assert_eq!(some_bool.compare_exchange(true,
                                      false,
                                      Ordering::Acquire,
                                      Ordering::Relaxed),
           Ok(true));
assert_eq!(some_bool.load(Ordering::Relaxed), false);

assert_eq!(some_bool.compare_exchange(true, true,
                                      Ordering::SeqCst,
                                      Ordering::Acquire),
           Err(false));
assert_eq!(some_bool.load(Ordering::Relaxed), false);

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim. This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. In this case, compare_exchange can lead to the ABA problem.

pub fn compare_exchange_weak( &self, current: bool, new: bool, success: Ordering, failure: Ordering, ) -> Result<bool, bool>

Stores a value into the bool if the current value is the same as the current value.

Unlike AtomicBool::compare_exchange, this function is allowed to spuriously fail even when the comparison succeeds, which can result in more efficient code on some platforms. The return value is a result indicating whether the new value was written and containing the previous value.

compare_exchange_weak takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicBool, Ordering};

let val = AtomicBool::new(false);

let new = true;
let mut old = val.load(Ordering::Relaxed);
loop {
    match val.compare_exchange_weak(old, new, Ordering::SeqCst, Ordering::Relaxed) {
        Ok(_) => break,
        Err(x) => old = x,
    }
}

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim. This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. In this case, compare_exchange can lead to the ABA problem.

pub fn fetch_and(&self, val: bool, order: Ordering) -> bool

Logical “and” with a boolean value.

Performs a logical “and” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_and takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicBool, Ordering};

let foo = AtomicBool::new(true);
assert_eq!(foo.fetch_and(false, Ordering::SeqCst), true);
assert_eq!(foo.load(Ordering::SeqCst), false);

let foo = AtomicBool::new(true);
assert_eq!(foo.fetch_and(true, Ordering::SeqCst), true);
assert_eq!(foo.load(Ordering::SeqCst), true);

let foo = AtomicBool::new(false);
assert_eq!(foo.fetch_and(false, Ordering::SeqCst), false);
assert_eq!(foo.load(Ordering::SeqCst), false);

pub fn fetch_nand(&self, val: bool, order: Ordering) -> bool

Logical “nand” with a boolean value.

Performs a logical “nand” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_nand takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicBool, Ordering};

let foo = AtomicBool::new(true);
assert_eq!(foo.fetch_nand(false, Ordering::SeqCst), true);
assert_eq!(foo.load(Ordering::SeqCst), true);

let foo = AtomicBool::new(true);
assert_eq!(foo.fetch_nand(true, Ordering::SeqCst), true);
assert_eq!(foo.load(Ordering::SeqCst) as usize, 0);
assert_eq!(foo.load(Ordering::SeqCst), false);

let foo = AtomicBool::new(false);
assert_eq!(foo.fetch_nand(false, Ordering::SeqCst), false);
assert_eq!(foo.load(Ordering::SeqCst), true);

pub fn fetch_or(&self, val: bool, order: Ordering) -> bool

Logical “or” with a boolean value.

Performs a logical “or” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_or takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicBool, Ordering};

let foo = AtomicBool::new(true);
assert_eq!(foo.fetch_or(false, Ordering::SeqCst), true);
assert_eq!(foo.load(Ordering::SeqCst), true);

let foo = AtomicBool::new(false);
assert_eq!(foo.fetch_or(true, Ordering::SeqCst), false);
assert_eq!(foo.load(Ordering::SeqCst), true);

let foo = AtomicBool::new(false);
assert_eq!(foo.fetch_or(false, Ordering::SeqCst), false);
assert_eq!(foo.load(Ordering::SeqCst), false);

pub fn fetch_xor(&self, val: bool, order: Ordering) -> bool

Logical “xor” with a boolean value.

Performs a logical “xor” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_xor takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicBool, Ordering};

let foo = AtomicBool::new(true);
assert_eq!(foo.fetch_xor(false, Ordering::SeqCst), true);
assert_eq!(foo.load(Ordering::SeqCst), true);

let foo = AtomicBool::new(true);
assert_eq!(foo.fetch_xor(true, Ordering::SeqCst), true);
assert_eq!(foo.load(Ordering::SeqCst), false);

let foo = AtomicBool::new(false);
assert_eq!(foo.fetch_xor(false, Ordering::SeqCst), false);
assert_eq!(foo.load(Ordering::SeqCst), false);

pub fn fetch_not(&self, order: Ordering) -> bool

Logical “not” with a boolean value.

Performs a logical “not” operation on the current value, and sets the new value to the result.

Returns the previous value.

fetch_not takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicBool, Ordering};

let foo = AtomicBool::new(true);
assert_eq!(foo.fetch_not(Ordering::SeqCst), true);
assert_eq!(foo.load(Ordering::SeqCst), false);

let foo = AtomicBool::new(false);
assert_eq!(foo.fetch_not(Ordering::SeqCst), false);
assert_eq!(foo.load(Ordering::SeqCst), true);

pub const fn as_ptr(&self) -> *mut bool

Returns a mutable pointer to the underlying bool.

Doing non-atomic reads and writes on the resulting boolean can be a data race. This method is mostly useful for FFI, where the function signature may use *mut bool instead of &AtomicBool.

Returning an *mut pointer from a shared reference to this atomic is safe because the atomic types work with interior mutability. All modifications of an atomic change the value through a shared reference, and can do so safely as long as they use atomic operations. Any use of the returned raw pointer requires an unsafe block and still has to uphold the requirements of the memory model.

Examples

use std::sync::atomic::AtomicBool;

extern "C" {
    fn my_atomic_op(arg: *mut bool);
}

let mut atomic = AtomicBool::new(true);
unsafe {
    my_atomic_op(atomic.as_ptr());
}

pub fn fetch_update<F>( &self, set_order: Ordering, fetch_order: Ordering, f: F, ) -> Result<bool, bool>

where F: FnMut(bool) -> Option<bool>,

👎Deprecating in 1.99.0: renamed to try_update for consistency

An alias for AtomicBool::try_update.

pub fn try_update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(bool) -> Option<bool>, ) -> Result<bool, bool>

Fetches the value, and applies a function to it that returns an optional new value. Returns a Result of Ok(previous_value) if the function returned Some(_), else Err(previous_value).

See also: update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, as long as the function returns Some(_), but the function will have been applied only once to the stored value.

try_update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicBool::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem.

Examples

use std::sync::atomic::{AtomicBool, Ordering};

let x = AtomicBool::new(false);
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |_| None), Err(false));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(!x)), Ok(false));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(!x)), Ok(true));
assert_eq!(x.load(Ordering::SeqCst), false);

pub fn update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(bool) -> bool, ) -> bool

Fetches the value, applies a function to it that it return a new value. The new value is stored and the old value is returned.

See also: try_update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, but the function will have been applied only once to the stored value.

update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicBool::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem.

Examples

use std::sync::atomic::{AtomicBool, Ordering};

let x = AtomicBool::new(false);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| !x), false);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| !x), true);
assert_eq!(x.load(Ordering::SeqCst), false);

impl<T> Atomic<*mut T>

pub const fn new(p: *mut T) -> AtomicPtr<T>

Creates a new AtomicPtr.

Examples

use std::sync::atomic::AtomicPtr;

let ptr = &mut 5;
let atomic_ptr = AtomicPtr::new(ptr);

pub const unsafe fn from_ptr<'a>(ptr: *mut *mut T) -> &'a AtomicPtr<T>

Creates a new AtomicPtr from a pointer.

Examples

use std::sync::atomic::{self, AtomicPtr};

// Get a pointer to an allocated value
let ptr: *mut *mut u8 = Box::into_raw(Box::new(std::ptr::null_mut()));

assert!(ptr.cast::<AtomicPtr<u8>>().is_aligned());

{
    // Create an atomic view of the allocated value
    let atomic = unsafe { AtomicPtr::from_ptr(ptr) };

    // Use `atomic` for atomic operations, possibly share it with other threads
    atomic.store(std::ptr::NonNull::dangling().as_ptr(), atomic::Ordering::Relaxed);
}

// It's ok to non-atomically access the value behind `ptr`,
// since the reference to the atomic ended its lifetime in the block above
assert!(!unsafe { *ptr }.is_null());

// Deallocate the value
unsafe { drop(Box::from_raw(ptr)) }

Safety

• ptr must be aligned to align_of::<AtomicPtr<T>>() (note that on some platforms this can be bigger than align_of::<*mut T>()).
• ptr must be valid for both reads and writes for the whole lifetime 'a.
• You must adhere to the Memory model for atomic accesses. In particular, it is not allowed to mix conflicting atomic and non-atomic accesses, or atomic accesses of different sizes, without synchronization.

pub const fn null() -> AtomicPtr<T>

🔬This is a nightly-only experimental API. (atomic_ptr_null #150733)

Creates a new AtomicPtr initialized with a null pointer.

Examples

#![feature(atomic_ptr_null)]
use std::sync::atomic::{AtomicPtr, Ordering};

let atomic_ptr = AtomicPtr::<()>::null();
assert!(atomic_ptr.load(Ordering::Relaxed).is_null());

pub fn get_mut(&mut self) -> &mut *mut T

Returns a mutable reference to the underlying pointer.

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::{AtomicPtr, Ordering};

let mut data = 10;
let mut atomic_ptr = AtomicPtr::new(&mut data);
let mut other_data = 5;
*atomic_ptr.get_mut() = &mut other_data;
assert_eq!(unsafe { *atomic_ptr.load(Ordering::SeqCst) }, 5);

pub fn from_mut(v: &mut *mut T) -> &mut Self

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Gets atomic access to a pointer.

Note: This function is only available on targets where AtomicPtr<T> has the same alignment as *const T

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicPtr, Ordering};

let mut data = 123;
let mut some_ptr = &mut data as *mut i32;
let a = AtomicPtr::from_mut(&mut some_ptr);
let mut other_data = 456;
a.store(&mut other_data, Ordering::Relaxed);
assert_eq!(unsafe { *some_ptr }, 456);

pub fn get_mut_slice(this: &mut [Self]) -> &mut [*mut T]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Gets non-atomic access to a &mut [AtomicPtr] slice.

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

#![feature(atomic_from_mut)]
use std::ptr::null_mut;
use std::sync::atomic::{AtomicPtr, Ordering};

let mut some_ptrs = [const { AtomicPtr::new(null_mut::<String>()) }; 10];

let view: &mut [*mut String] = AtomicPtr::get_mut_slice(&mut some_ptrs);
assert_eq!(view, [null_mut::<String>(); 10]);
view
    .iter_mut()
    .enumerate()
    .for_each(|(i, ptr)| *ptr = Box::into_raw(Box::new(format!("iteration#{i}"))));

std::thread::scope(|s| {
    for ptr in &some_ptrs {
        s.spawn(move || {
            let ptr = ptr.load(Ordering::Relaxed);
            assert!(!ptr.is_null());

            let name = unsafe { Box::from_raw(ptr) };
            println!("Hello, {name}!");
        });
    }
});

pub fn from_mut_slice(v: &mut [*mut T]) -> &mut [Self]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Gets atomic access to a slice of pointers.

Note: This function is only available on targets where AtomicPtr<T> has the same alignment as *const T

Examples

#![feature(atomic_from_mut)]
use std::ptr::null_mut;
use std::sync::atomic::{AtomicPtr, Ordering};

let mut some_ptrs = [null_mut::<String>(); 10];
let a = &*AtomicPtr::from_mut_slice(&mut some_ptrs);
std::thread::scope(|s| {
    for i in 0..a.len() {
        s.spawn(move || {
            let name = Box::new(format!("thread{i}"));
            a[i].store(Box::into_raw(name), Ordering::Relaxed);
        });
    }
});
for p in some_ptrs {
    assert!(!p.is_null());
    let name = unsafe { Box::from_raw(p) };
    println!("Hello, {name}!");
}

pub const fn into_inner(self) -> *mut T

Consumes the atomic and returns the contained value.

This is safe because passing self by value guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::AtomicPtr;

let mut data = 5;
let atomic_ptr = AtomicPtr::new(&mut data);
assert_eq!(unsafe { *atomic_ptr.into_inner() }, 5);

pub fn load(&self, order: Ordering) -> *mut T

Loads a value from the pointer.

load takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Acquire and Relaxed.

Panics

Panics if order is Release or AcqRel.

Examples

use std::sync::atomic::{AtomicPtr, Ordering};

let ptr = &mut 5;
let some_ptr = AtomicPtr::new(ptr);

let value = some_ptr.load(Ordering::Relaxed);

pub fn store(&self, ptr: *mut T, order: Ordering)

Stores a value into the pointer.

store takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Release and Relaxed.

Panics

Panics if order is Acquire or AcqRel.

Examples

use std::sync::atomic::{AtomicPtr, Ordering};

let ptr = &mut 5;
let some_ptr = AtomicPtr::new(ptr);

let other_ptr = &mut 10;

some_ptr.store(other_ptr, Ordering::Relaxed);

pub fn swap(&self, ptr: *mut T, order: Ordering) -> *mut T

Stores a value into the pointer, returning the previous value.

swap takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on pointers.

Examples

use std::sync::atomic::{AtomicPtr, Ordering};

let ptr = &mut 5;
let some_ptr = AtomicPtr::new(ptr);

let other_ptr = &mut 10;

let value = some_ptr.swap(other_ptr, Ordering::Relaxed);

pub fn compare_and_swap( &self, current: *mut T, new: *mut T, order: Ordering, ) -> *mut T

👎Deprecated since 1.50.0: Use compare_exchange or compare_exchange_weak instead

Stores a value into the pointer if the current value is the same as the current value.

The return value is always the previous value. If it is equal to current, then the value was updated.

compare_and_swap also takes an Ordering argument which describes the memory ordering of this operation. Notice that even when using AcqRel, the operation might fail and hence just perform an Acquire load, but not have Release semantics. Using Acquire makes the store part of this operation Relaxed if it happens, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on pointers.

Migrating to compare_exchange and compare_exchange_weak

compare_and_swap is equivalent to compare_exchange with the following mapping for memory orderings:

Original	Success	Failure
Relaxed	Relaxed	Relaxed
Acquire	Acquire	Acquire
Release	Release	Relaxed
AcqRel	AcqRel	Acquire
SeqCst	SeqCst	SeqCst

compare_and_swap and compare_exchange also differ in their return type. You can use compare_exchange(...).unwrap_or_else(|x| x) to recover the behavior of compare_and_swap, but in most cases it is more idiomatic to check whether the return value is Ok or Err rather than to infer success vs failure based on the value that was read.

During migration, consider whether it makes sense to use compare_exchange_weak instead. compare_exchange_weak is allowed to fail spuriously even when the comparison succeeds, which allows the compiler to generate better assembly code when the compare and swap is used in a loop.

Examples

use std::sync::atomic::{AtomicPtr, Ordering};

let ptr = &mut 5;
let some_ptr = AtomicPtr::new(ptr);

let other_ptr = &mut 10;

let value = some_ptr.compare_and_swap(ptr, other_ptr, Ordering::Relaxed);

pub fn compare_exchange( &self, current: *mut T, new: *mut T, success: Ordering, failure: Ordering, ) -> Result<*mut T, *mut T>

Stores a value into the pointer if the current value is the same as the current value.

The return value is a result indicating whether the new value was written and containing the previous value. On success this value is guaranteed to be equal to current.

compare_exchange takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on pointers.

Examples

use std::sync::atomic::{AtomicPtr, Ordering};

let ptr = &mut 5;
let some_ptr = AtomicPtr::new(ptr);

let other_ptr = &mut 10;

let value = some_ptr.compare_exchange(ptr, other_ptr,
                                      Ordering::SeqCst, Ordering::Relaxed);

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim. This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn compare_exchange_weak( &self, current: *mut T, new: *mut T, success: Ordering, failure: Ordering, ) -> Result<*mut T, *mut T>

Stores a value into the pointer if the current value is the same as the current value.

Unlike AtomicPtr::compare_exchange, this function is allowed to spuriously fail even when the comparison succeeds, which can result in more efficient code on some platforms. The return value is a result indicating whether the new value was written and containing the previous value.

compare_exchange_weak takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on pointers.

Examples

use std::sync::atomic::{AtomicPtr, Ordering};

let some_ptr = AtomicPtr::new(&mut 5);

let new = &mut 10;
let mut old = some_ptr.load(Ordering::Relaxed);
loop {
    match some_ptr.compare_exchange_weak(old, new, Ordering::SeqCst, Ordering::Relaxed) {
        Ok(_) => break,
        Err(x) => old = x,
    }
}

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim. This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn fetch_update<F>( &self, set_order: Ordering, fetch_order: Ordering, f: F, ) -> Result<*mut T, *mut T>

where F: FnMut(*mut T) -> Option<*mut T>,

👎Deprecating in 1.99.0: renamed to try_update for consistency

An alias for AtomicPtr::try_update.

pub fn try_update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(*mut T) -> Option<*mut T>, ) -> Result<*mut T, *mut T>

Fetches the value, and applies a function to it that returns an optional new value. Returns a Result of Ok(previous_value) if the function returned Some(_), else Err(previous_value).

See also: update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, as long as the function returns Some(_), but the function will have been applied only once to the stored value.

try_update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicPtr::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on pointers.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem, which is a particularly common pitfall for pointers!

Examples

use std::sync::atomic::{AtomicPtr, Ordering};

let ptr: *mut _ = &mut 5;
let some_ptr = AtomicPtr::new(ptr);

let new: *mut _ = &mut 10;
assert_eq!(some_ptr.try_update(Ordering::SeqCst, Ordering::SeqCst, |_| None), Err(ptr));
let result = some_ptr.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| {
    if x == ptr {
        Some(new)
    } else {
        None
    }
});
assert_eq!(result, Ok(ptr));
assert_eq!(some_ptr.load(Ordering::SeqCst), new);

pub fn update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(*mut T) -> *mut T, ) -> *mut T

Fetches the value, applies a function to it that it return a new value. The new value is stored and the old value is returned.

See also: try_update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, but the function will have been applied only once to the stored value.

update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicPtr::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on pointers.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem, which is a particularly common pitfall for pointers!

Examples

use std::sync::atomic::{AtomicPtr, Ordering};

let ptr: *mut _ = &mut 5;
let some_ptr = AtomicPtr::new(ptr);

let new: *mut _ = &mut 10;
let result = some_ptr.update(Ordering::SeqCst, Ordering::SeqCst, |_| new);
assert_eq!(result, ptr);
assert_eq!(some_ptr.load(Ordering::SeqCst), new);

pub fn fetch_ptr_add(&self, val: usize, order: Ordering) -> *mut T

Offsets the pointer’s address by adding val (in units of T), returning the previous pointer.

This is equivalent to using wrapping_add to atomically perform the equivalent of ptr = ptr.wrapping_add(val);.

This method operates in units of T, which means that it cannot be used to offset the pointer by an amount which is not a multiple of size_of::<T>(). This can sometimes be inconvenient, as you may want to work with a deliberately misaligned pointer. In such cases, you may use the fetch_byte_add method instead.

fetch_ptr_add takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on AtomicPtr.

Examples

use core::sync::atomic::{AtomicPtr, Ordering};

let atom = AtomicPtr::<i64>::new(core::ptr::null_mut());
assert_eq!(atom.fetch_ptr_add(1, Ordering::Relaxed).addr(), 0);
// Note: units of `size_of::<i64>()`.
assert_eq!(atom.load(Ordering::Relaxed).addr(), 8);

pub fn fetch_ptr_sub(&self, val: usize, order: Ordering) -> *mut T

Offsets the pointer’s address by subtracting val (in units of T), returning the previous pointer.

This is equivalent to using wrapping_sub to atomically perform the equivalent of ptr = ptr.wrapping_sub(val);.

This method operates in units of T, which means that it cannot be used to offset the pointer by an amount which is not a multiple of size_of::<T>(). This can sometimes be inconvenient, as you may want to work with a deliberately misaligned pointer. In such cases, you may use the fetch_byte_sub method instead.

fetch_ptr_sub takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on AtomicPtr.

Examples

use core::sync::atomic::{AtomicPtr, Ordering};

let array = [1i32, 2i32];
let atom = AtomicPtr::new(array.as_ptr().wrapping_add(1) as *mut _);

assert!(core::ptr::eq(
    atom.fetch_ptr_sub(1, Ordering::Relaxed),
    &array[1],
));
assert!(core::ptr::eq(atom.load(Ordering::Relaxed), &array[0]));

pub fn fetch_byte_add(&self, val: usize, order: Ordering) -> *mut T

Offsets the pointer’s address by adding val bytes, returning the previous pointer.

This is equivalent to using wrapping_byte_add to atomically perform ptr = ptr.wrapping_byte_add(val).

fetch_byte_add takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on AtomicPtr.

Examples

use core::sync::atomic::{AtomicPtr, Ordering};

let atom = AtomicPtr::<i64>::new(core::ptr::null_mut());
assert_eq!(atom.fetch_byte_add(1, Ordering::Relaxed).addr(), 0);
// Note: in units of bytes, not `size_of::<i64>()`.
assert_eq!(atom.load(Ordering::Relaxed).addr(), 1);

pub fn fetch_byte_sub(&self, val: usize, order: Ordering) -> *mut T

Offsets the pointer’s address by subtracting val bytes, returning the previous pointer.

This is equivalent to using wrapping_byte_sub to atomically perform ptr = ptr.wrapping_byte_sub(val).

fetch_byte_sub takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on AtomicPtr.

Examples

use core::sync::atomic::{AtomicPtr, Ordering};

let mut arr = [0i64, 1];
let atom = AtomicPtr::<i64>::new(&raw mut arr[1]);
assert_eq!(atom.fetch_byte_sub(8, Ordering::Relaxed).addr(), (&raw const arr[1]).addr());
assert_eq!(atom.load(Ordering::Relaxed).addr(), (&raw const arr[0]).addr());

pub fn fetch_or(&self, val: usize, order: Ordering) -> *mut T

Performs a bitwise “or” operation on the address of the current pointer, and the argument val, and stores a pointer with provenance of the current pointer and the resulting address.

This is equivalent to using map_addr to atomically perform ptr = ptr.map_addr(|a| a | val). This can be used in tagged pointer schemes to atomically set tag bits.

Caveat: This operation returns the previous value. To compute the stored value without losing provenance, you may use map_addr. For example: a.fetch_or(val).map_addr(|a| a | val).

fetch_or takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on AtomicPtr.

This API and its claimed semantics are part of the Strict Provenance experiment, see the module documentation for ptr for details.

Examples

use core::sync::atomic::{AtomicPtr, Ordering};

let pointer = &mut 3i64 as *mut i64;

let atom = AtomicPtr::<i64>::new(pointer);
// Tag the bottom bit of the pointer.
assert_eq!(atom.fetch_or(1, Ordering::Relaxed).addr() & 1, 0);
// Extract and untag.
let tagged = atom.load(Ordering::Relaxed);
assert_eq!(tagged.addr() & 1, 1);
assert_eq!(tagged.map_addr(|p| p & !1), pointer);

pub fn fetch_and(&self, val: usize, order: Ordering) -> *mut T

Performs a bitwise “and” operation on the address of the current pointer, and the argument val, and stores a pointer with provenance of the current pointer and the resulting address.

This is equivalent to using map_addr to atomically perform ptr = ptr.map_addr(|a| a & val). This can be used in tagged pointer schemes to atomically unset tag bits.

Caveat: This operation returns the previous value. To compute the stored value without losing provenance, you may use map_addr. For example: a.fetch_and(val).map_addr(|a| a & val).

fetch_and takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on AtomicPtr.

This API and its claimed semantics are part of the Strict Provenance experiment, see the module documentation for ptr for details.

Examples

use core::sync::atomic::{AtomicPtr, Ordering};

let pointer = &mut 3i64 as *mut i64;
// A tagged pointer
let atom = AtomicPtr::<i64>::new(pointer.map_addr(|a| a | 1));
assert_eq!(atom.fetch_or(1, Ordering::Relaxed).addr() & 1, 1);
// Untag, and extract the previously tagged pointer.
let untagged = atom.fetch_and(!1, Ordering::Relaxed)
    .map_addr(|a| a & !1);
assert_eq!(untagged, pointer);

pub fn fetch_xor(&self, val: usize, order: Ordering) -> *mut T

Performs a bitwise “xor” operation on the address of the current pointer, and the argument val, and stores a pointer with provenance of the current pointer and the resulting address.

This is equivalent to using map_addr to atomically perform ptr = ptr.map_addr(|a| a ^ val). This can be used in tagged pointer schemes to atomically toggle tag bits.

Caveat: This operation returns the previous value. To compute the stored value without losing provenance, you may use map_addr. For example: a.fetch_xor(val).map_addr(|a| a ^ val).

fetch_xor takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on AtomicPtr.

This API and its claimed semantics are part of the Strict Provenance experiment, see the module documentation for ptr for details.

Examples

use core::sync::atomic::{AtomicPtr, Ordering};

let pointer = &mut 3i64 as *mut i64;
let atom = AtomicPtr::<i64>::new(pointer);

// Toggle a tag bit on the pointer.
atom.fetch_xor(1, Ordering::Relaxed);
assert_eq!(atom.load(Ordering::Relaxed).addr() & 1, 1);

pub const fn as_ptr(&self) -> *mut *mut T

Returns a mutable pointer to the underlying pointer.

Doing non-atomic reads and writes on the resulting pointer can be a data race. This method is mostly useful for FFI, where the function signature may use *mut *mut T instead of &AtomicPtr<T>.

Returning an *mut pointer from a shared reference to this atomic is safe because the atomic types work with interior mutability. All modifications of an atomic change the value through a shared reference, and can do so safely as long as they use atomic operations. Any use of the returned raw pointer requires an unsafe block and still has to uphold the requirements of the memory model.

Examples

use std::sync::atomic::AtomicPtr;

extern "C" {
    fn my_atomic_op(arg: *mut *mut u32);
}

let mut value = 17;
let atomic = AtomicPtr::new(&mut value);

// SAFETY: Safe as long as `my_atomic_op` is atomic.
unsafe {
    my_atomic_op(atomic.as_ptr());
}

impl Atomic<i8>

pub const fn new(v: i8) -> Self

Creates a new atomic integer.

Examples

use std::sync::atomic::AtomicI8;

let atomic_forty_two = AtomicI8::new(42);

pub const unsafe fn from_ptr<'a>(ptr: *mut i8) -> &'a AtomicI8

Creates a new reference to an atomic integer from a pointer.

Examples

use std::sync::atomic::{self, AtomicI8};

// Get a pointer to an allocated value
let ptr: *mut i8 = Box::into_raw(Box::new(0));

assert!(ptr.cast::<AtomicI8>().is_aligned());

{
    // Create an atomic view of the allocated value
    let atomic = unsafe {AtomicI8::from_ptr(ptr) };

    // Use `atomic` for atomic operations, possibly share it with other threads
    atomic.store(1, atomic::Ordering::Relaxed);
}

// It's ok to non-atomically access the value behind `ptr`,
// since the reference to the atomic ended its lifetime in the block above
assert_eq!(unsafe { *ptr }, 1);

// Deallocate the value
unsafe { drop(Box::from_raw(ptr)) }

Safety

• ptr must be aligned to align_of::<AtomicI8>() (note that this is always true, since align_of::<AtomicI8>() == 1).
• ptr must be valid for both reads and writes for the whole lifetime 'a.
• You must adhere to the Memory model for atomic accesses. In particular, it is not allowed to mix conflicting atomic and non-atomic accesses, or atomic accesses of different sizes, without synchronization.

pub fn get_mut(&mut self) -> &mut i8

Returns a mutable reference to the underlying integer.

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let mut some_var = AtomicI8::new(10);
assert_eq!(*some_var.get_mut(), 10);
*some_var.get_mut() = 5;
assert_eq!(some_var.load(Ordering::SeqCst), 5);

pub fn from_mut(v: &mut i8) -> &mut Self

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut i8.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicI8, Ordering};

let mut some_int = 123;
let a = AtomicI8::from_mut(&mut some_int);
a.store(100, Ordering::Relaxed);
assert_eq!(some_int, 100);

pub fn get_mut_slice(this: &mut [Self]) -> &mut [i8]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get non-atomic access to a &mut [AtomicI8] slice

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicI8, Ordering};

let mut some_ints = [const { AtomicI8::new(0) }; 10];

let view: &mut [i8] = AtomicI8::get_mut_slice(&mut some_ints);
assert_eq!(view, [0; 10]);
view
    .iter_mut()
    .enumerate()
    .for_each(|(idx, int)| *int = idx as _);

std::thread::scope(|s| {
    some_ints
        .iter()
        .enumerate()
        .for_each(|(idx, int)| {
            s.spawn(move || assert_eq!(int.load(Ordering::Relaxed), idx as _));
        })
});

pub fn from_mut_slice(v: &mut [i8]) -> &mut [Self]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut [i8] slice.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicI8, Ordering};

let mut some_ints = [0; 10];
let a = &*AtomicI8::from_mut_slice(&mut some_ints);
std::thread::scope(|s| {
    for i in 0..a.len() {
        s.spawn(move || a[i].store(i as _, Ordering::Relaxed));
    }
});
for (i, n) in some_ints.into_iter().enumerate() {
    assert_eq!(i, n as usize);
}

pub const fn into_inner(self) -> i8

Consumes the atomic and returns the contained value.

This is safe because passing self by value guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::AtomicI8;

let some_var = AtomicI8::new(5);
assert_eq!(some_var.into_inner(), 5);

pub fn load(&self, order: Ordering) -> i8

Loads a value from the atomic integer.

load takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Acquire and Relaxed.

Panics

Panics if order is Release or AcqRel.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let some_var = AtomicI8::new(5);

assert_eq!(some_var.load(Ordering::Relaxed), 5);

pub fn store(&self, val: i8, order: Ordering)

Stores a value into the atomic integer.

store takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Release and Relaxed.

Panics

Panics if order is Acquire or AcqRel.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let some_var = AtomicI8::new(5);

some_var.store(10, Ordering::Relaxed);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn swap(&self, val: i8, order: Ordering) -> i8

Stores a value into the atomic integer, returning the previous value.

swap takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i8.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let some_var = AtomicI8::new(5);

assert_eq!(some_var.swap(10, Ordering::Relaxed), 5);

pub fn compare_and_swap(&self, current: i8, new: i8, order: Ordering) -> i8

👎Deprecated since 1.50.0: Use compare_exchange or compare_exchange_weak instead

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is always the previous value. If it is equal to current, then the value was updated.

compare_and_swap also takes an Ordering argument which describes the memory ordering of this operation. Notice that even when using AcqRel, the operation might fail and hence just perform an Acquire load, but not have Release semantics. Using Acquire makes the store part of this operation Relaxed if it happens, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i8.

Migrating to compare_exchange and compare_exchange_weak

compare_and_swap is equivalent to compare_exchange with the following mapping for memory orderings:

Original	Success	Failure
Relaxed	Relaxed	Relaxed
Acquire	Acquire	Acquire
Release	Release	Relaxed
AcqRel	AcqRel	Acquire
SeqCst	SeqCst	SeqCst

compare_and_swap and compare_exchange also differ in their return type. You can use compare_exchange(...).unwrap_or_else(|x| x) to recover the behavior of compare_and_swap, but in most cases it is more idiomatic to check whether the return value is Ok or Err rather than to infer success vs failure based on the value that was read.

During migration, consider whether it makes sense to use compare_exchange_weak instead. compare_exchange_weak is allowed to fail spuriously even when the comparison succeeds, which allows the compiler to generate better assembly code when the compare and swap is used in a loop.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let some_var = AtomicI8::new(5);

assert_eq!(some_var.compare_and_swap(5, 10, Ordering::Relaxed), 5);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_and_swap(6, 12, Ordering::Relaxed), 10);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn compare_exchange( &self, current: i8, new: i8, success: Ordering, failure: Ordering, ) -> Result<i8, i8>

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is a result indicating whether the new value was written and containing the previous value. On success this value is guaranteed to be equal to current.

compare_exchange takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on i8.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let some_var = AtomicI8::new(5);

assert_eq!(some_var.compare_exchange(5, 10,
                                     Ordering::Acquire,
                                     Ordering::Relaxed),
           Ok(5));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_exchange(6, 12,
                                     Ordering::SeqCst,
                                     Ordering::Acquire),
           Err(10));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim! This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn compare_exchange_weak( &self, current: i8, new: i8, success: Ordering, failure: Ordering, ) -> Result<i8, i8>

Stores a value into the atomic integer if the current value is the same as the current value.

Unlike AtomicI8::compare_exchange, this function is allowed to spuriously fail even when the comparison succeeds, which can result in more efficient code on some platforms. The return value is a result indicating whether the new value was written and containing the previous value.

compare_exchange_weak takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on i8.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let val = AtomicI8::new(4);

let mut old = val.load(Ordering::Relaxed);
loop {
    let new = old * 2;
    match val.compare_exchange_weak(old, new, Ordering::SeqCst, Ordering::Relaxed) {
        Ok(_) => break,
        Err(x) => old = x,
    }
}

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim. This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn fetch_add(&self, val: i8, order: Ordering) -> i8

Adds to the current value, returning the previous value.

This operation wraps around on overflow.

fetch_add takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i8.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let foo = AtomicI8::new(0);
assert_eq!(foo.fetch_add(10, Ordering::SeqCst), 0);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_sub(&self, val: i8, order: Ordering) -> i8

Subtracts from the current value, returning the previous value.

This operation wraps around on overflow.

fetch_sub takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i8.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let foo = AtomicI8::new(20);
assert_eq!(foo.fetch_sub(10, Ordering::SeqCst), 20);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_and(&self, val: i8, order: Ordering) -> i8

Bitwise “and” with the current value.

Performs a bitwise “and” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_and takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i8.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let foo = AtomicI8::new(0b101101);
assert_eq!(foo.fetch_and(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b100001);

pub fn fetch_nand(&self, val: i8, order: Ordering) -> i8

Bitwise “nand” with the current value.

Performs a bitwise “nand” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_nand takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i8.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let foo = AtomicI8::new(0x13);
assert_eq!(foo.fetch_nand(0x31, Ordering::SeqCst), 0x13);
assert_eq!(foo.load(Ordering::SeqCst), !(0x13 & 0x31));

pub fn fetch_or(&self, val: i8, order: Ordering) -> i8

Bitwise “or” with the current value.

Performs a bitwise “or” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_or takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i8.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let foo = AtomicI8::new(0b101101);
assert_eq!(foo.fetch_or(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b111111);

pub fn fetch_xor(&self, val: i8, order: Ordering) -> i8

Bitwise “xor” with the current value.

Performs a bitwise “xor” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_xor takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i8.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let foo = AtomicI8::new(0b101101);
assert_eq!(foo.fetch_xor(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b011110);

pub fn fetch_update<F>( &self, set_order: Ordering, fetch_order: Ordering, f: F, ) -> Result<i8, i8>

where F: FnMut(i8) -> Option<i8>,

👎Deprecating in 1.99.0: renamed to try_update for consistency

An alias for AtomicI8::try_update .

pub fn try_update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(i8) -> Option<i8>, ) -> Result<i8, i8>

Fetches the value, and applies a function to it that returns an optional new value. Returns a Result of Ok(previous_value) if the function returned Some(_), else Err(previous_value).

See also: update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, as long as the function returns Some(_), but the function will have been applied only once to the stored value.

try_update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicI8::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on i8.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let x = AtomicI8::new(7);
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |_| None), Err(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(8));
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(i8) -> i8, ) -> i8

Fetches the value, applies a function to it that it return a new value. The new value is stored and the old value is returned.

See also: try_update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, but the function will have been applied only once to the stored value.

update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicI8::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on i8.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let x = AtomicI8::new(7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 8);
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn fetch_max(&self, val: i8, order: Ordering) -> i8

Maximum with the current value.

Finds the maximum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_max takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i8.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let foo = AtomicI8::new(23);
assert_eq!(foo.fetch_max(42, Ordering::SeqCst), 23);
assert_eq!(foo.load(Ordering::SeqCst), 42);

If you want to obtain the maximum value in one step, you can use the following:

use std::sync::atomic::{AtomicI8, Ordering};

let foo = AtomicI8::new(23);
let bar = 42;
let max_foo = foo.fetch_max(bar, Ordering::SeqCst).max(bar);
assert!(max_foo == 42);

pub fn fetch_min(&self, val: i8, order: Ordering) -> i8

Minimum with the current value.

Finds the minimum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_min takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i8.

Examples

use std::sync::atomic::{AtomicI8, Ordering};

let foo = AtomicI8::new(23);
assert_eq!(foo.fetch_min(42, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 23);
assert_eq!(foo.fetch_min(22, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 22);

If you want to obtain the minimum value in one step, you can use the following:

use std::sync::atomic::{AtomicI8, Ordering};

let foo = AtomicI8::new(23);
let bar = 12;
let min_foo = foo.fetch_min(bar, Ordering::SeqCst).min(bar);
assert_eq!(min_foo, 12);

pub const fn as_ptr(&self) -> *mut i8

Returns a mutable pointer to the underlying integer.

Doing non-atomic reads and writes on the resulting integer can be a data race. This method is mostly useful for FFI, where the function signature may use *mut i8 instead of &AtomicI8.

Returning an *mut pointer from a shared reference to this atomic is safe because the atomic types work with interior mutability. All modifications of an atomic change the value through a shared reference, and can do so safely as long as they use atomic operations. Any use of the returned raw pointer requires an unsafe block and still has to uphold the requirements of the memory model.

Examples

use std::sync::atomic::AtomicI8;

extern "C" {
    fn my_atomic_op(arg: *mut i8);
}

let atomic = AtomicI8::new(1);

// SAFETY: Safe as long as `my_atomic_op` is atomic.
unsafe {
    my_atomic_op(atomic.as_ptr());
}

impl Atomic<u8>

pub const fn new(v: u8) -> Self

Creates a new atomic integer.

Examples

use std::sync::atomic::AtomicU8;

let atomic_forty_two = AtomicU8::new(42);

pub const unsafe fn from_ptr<'a>(ptr: *mut u8) -> &'a AtomicU8

Creates a new reference to an atomic integer from a pointer.

Examples

use std::sync::atomic::{self, AtomicU8};

// Get a pointer to an allocated value
let ptr: *mut u8 = Box::into_raw(Box::new(0));

assert!(ptr.cast::<AtomicU8>().is_aligned());

{
    // Create an atomic view of the allocated value
    let atomic = unsafe {AtomicU8::from_ptr(ptr) };

    // Use `atomic` for atomic operations, possibly share it with other threads
    atomic.store(1, atomic::Ordering::Relaxed);
}

// It's ok to non-atomically access the value behind `ptr`,
// since the reference to the atomic ended its lifetime in the block above
assert_eq!(unsafe { *ptr }, 1);

// Deallocate the value
unsafe { drop(Box::from_raw(ptr)) }

Safety

• ptr must be aligned to align_of::<AtomicU8>() (note that this is always true, since align_of::<AtomicU8>() == 1).
• ptr must be valid for both reads and writes for the whole lifetime 'a.
• You must adhere to the Memory model for atomic accesses. In particular, it is not allowed to mix conflicting atomic and non-atomic accesses, or atomic accesses of different sizes, without synchronization.

pub fn get_mut(&mut self) -> &mut u8

Returns a mutable reference to the underlying integer.

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let mut some_var = AtomicU8::new(10);
assert_eq!(*some_var.get_mut(), 10);
*some_var.get_mut() = 5;
assert_eq!(some_var.load(Ordering::SeqCst), 5);

pub fn from_mut(v: &mut u8) -> &mut Self

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut u8.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicU8, Ordering};

let mut some_int = 123;
let a = AtomicU8::from_mut(&mut some_int);
a.store(100, Ordering::Relaxed);
assert_eq!(some_int, 100);

pub fn get_mut_slice(this: &mut [Self]) -> &mut [u8]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get non-atomic access to a &mut [AtomicU8] slice

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicU8, Ordering};

let mut some_ints = [const { AtomicU8::new(0) }; 10];

let view: &mut [u8] = AtomicU8::get_mut_slice(&mut some_ints);
assert_eq!(view, [0; 10]);
view
    .iter_mut()
    .enumerate()
    .for_each(|(idx, int)| *int = idx as _);

std::thread::scope(|s| {
    some_ints
        .iter()
        .enumerate()
        .for_each(|(idx, int)| {
            s.spawn(move || assert_eq!(int.load(Ordering::Relaxed), idx as _));
        })
});

pub fn from_mut_slice(v: &mut [u8]) -> &mut [Self]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut [u8] slice.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicU8, Ordering};

let mut some_ints = [0; 10];
let a = &*AtomicU8::from_mut_slice(&mut some_ints);
std::thread::scope(|s| {
    for i in 0..a.len() {
        s.spawn(move || a[i].store(i as _, Ordering::Relaxed));
    }
});
for (i, n) in some_ints.into_iter().enumerate() {
    assert_eq!(i, n as usize);
}

pub const fn into_inner(self) -> u8

Consumes the atomic and returns the contained value.

This is safe because passing self by value guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::AtomicU8;

let some_var = AtomicU8::new(5);
assert_eq!(some_var.into_inner(), 5);

pub fn load(&self, order: Ordering) -> u8

Loads a value from the atomic integer.

load takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Acquire and Relaxed.

Panics

Panics if order is Release or AcqRel.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let some_var = AtomicU8::new(5);

assert_eq!(some_var.load(Ordering::Relaxed), 5);

pub fn store(&self, val: u8, order: Ordering)

Stores a value into the atomic integer.

store takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Release and Relaxed.

Panics

Panics if order is Acquire or AcqRel.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let some_var = AtomicU8::new(5);

some_var.store(10, Ordering::Relaxed);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn swap(&self, val: u8, order: Ordering) -> u8

Stores a value into the atomic integer, returning the previous value.

swap takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let some_var = AtomicU8::new(5);

assert_eq!(some_var.swap(10, Ordering::Relaxed), 5);

pub fn compare_and_swap(&self, current: u8, new: u8, order: Ordering) -> u8

👎Deprecated since 1.50.0: Use compare_exchange or compare_exchange_weak instead

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is always the previous value. If it is equal to current, then the value was updated.

compare_and_swap also takes an Ordering argument which describes the memory ordering of this operation. Notice that even when using AcqRel, the operation might fail and hence just perform an Acquire load, but not have Release semantics. Using Acquire makes the store part of this operation Relaxed if it happens, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Migrating to compare_exchange and compare_exchange_weak

compare_and_swap is equivalent to compare_exchange with the following mapping for memory orderings:

Original	Success	Failure
Relaxed	Relaxed	Relaxed
Acquire	Acquire	Acquire
Release	Release	Relaxed
AcqRel	AcqRel	Acquire
SeqCst	SeqCst	SeqCst

compare_and_swap and compare_exchange also differ in their return type. You can use compare_exchange(...).unwrap_or_else(|x| x) to recover the behavior of compare_and_swap, but in most cases it is more idiomatic to check whether the return value is Ok or Err rather than to infer success vs failure based on the value that was read.

During migration, consider whether it makes sense to use compare_exchange_weak instead. compare_exchange_weak is allowed to fail spuriously even when the comparison succeeds, which allows the compiler to generate better assembly code when the compare and swap is used in a loop.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let some_var = AtomicU8::new(5);

assert_eq!(some_var.compare_and_swap(5, 10, Ordering::Relaxed), 5);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_and_swap(6, 12, Ordering::Relaxed), 10);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn compare_exchange( &self, current: u8, new: u8, success: Ordering, failure: Ordering, ) -> Result<u8, u8>

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is a result indicating whether the new value was written and containing the previous value. On success this value is guaranteed to be equal to current.

compare_exchange takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let some_var = AtomicU8::new(5);

assert_eq!(some_var.compare_exchange(5, 10,
                                     Ordering::Acquire,
                                     Ordering::Relaxed),
           Ok(5));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_exchange(6, 12,
                                     Ordering::SeqCst,
                                     Ordering::Acquire),
           Err(10));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim! This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn compare_exchange_weak( &self, current: u8, new: u8, success: Ordering, failure: Ordering, ) -> Result<u8, u8>

Stores a value into the atomic integer if the current value is the same as the current value.

Unlike AtomicU8::compare_exchange, this function is allowed to spuriously fail even when the comparison succeeds, which can result in more efficient code on some platforms. The return value is a result indicating whether the new value was written and containing the previous value.

compare_exchange_weak takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let val = AtomicU8::new(4);

let mut old = val.load(Ordering::Relaxed);
loop {
    let new = old * 2;
    match val.compare_exchange_weak(old, new, Ordering::SeqCst, Ordering::Relaxed) {
        Ok(_) => break,
        Err(x) => old = x,
    }
}

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim. This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn fetch_add(&self, val: u8, order: Ordering) -> u8

Adds to the current value, returning the previous value.

This operation wraps around on overflow.

fetch_add takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let foo = AtomicU8::new(0);
assert_eq!(foo.fetch_add(10, Ordering::SeqCst), 0);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_sub(&self, val: u8, order: Ordering) -> u8

Subtracts from the current value, returning the previous value.

This operation wraps around on overflow.

fetch_sub takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let foo = AtomicU8::new(20);
assert_eq!(foo.fetch_sub(10, Ordering::SeqCst), 20);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_and(&self, val: u8, order: Ordering) -> u8

Bitwise “and” with the current value.

Performs a bitwise “and” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_and takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let foo = AtomicU8::new(0b101101);
assert_eq!(foo.fetch_and(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b100001);

pub fn fetch_nand(&self, val: u8, order: Ordering) -> u8

Bitwise “nand” with the current value.

Performs a bitwise “nand” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_nand takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let foo = AtomicU8::new(0x13);
assert_eq!(foo.fetch_nand(0x31, Ordering::SeqCst), 0x13);
assert_eq!(foo.load(Ordering::SeqCst), !(0x13 & 0x31));

pub fn fetch_or(&self, val: u8, order: Ordering) -> u8

Bitwise “or” with the current value.

Performs a bitwise “or” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_or takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let foo = AtomicU8::new(0b101101);
assert_eq!(foo.fetch_or(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b111111);

pub fn fetch_xor(&self, val: u8, order: Ordering) -> u8

Bitwise “xor” with the current value.

Performs a bitwise “xor” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_xor takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let foo = AtomicU8::new(0b101101);
assert_eq!(foo.fetch_xor(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b011110);

pub fn fetch_update<F>( &self, set_order: Ordering, fetch_order: Ordering, f: F, ) -> Result<u8, u8>

where F: FnMut(u8) -> Option<u8>,

👎Deprecating in 1.99.0: renamed to try_update for consistency

An alias for AtomicU8::try_update .

pub fn try_update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(u8) -> Option<u8>, ) -> Result<u8, u8>

Fetches the value, and applies a function to it that returns an optional new value. Returns a Result of Ok(previous_value) if the function returned Some(_), else Err(previous_value).

See also: update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, as long as the function returns Some(_), but the function will have been applied only once to the stored value.

try_update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicU8::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let x = AtomicU8::new(7);
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |_| None), Err(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(8));
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(u8) -> u8, ) -> u8

Fetches the value, applies a function to it that it return a new value. The new value is stored and the old value is returned.

See also: try_update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, but the function will have been applied only once to the stored value.

update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicU8::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let x = AtomicU8::new(7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 8);
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn fetch_max(&self, val: u8, order: Ordering) -> u8

Maximum with the current value.

Finds the maximum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_max takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let foo = AtomicU8::new(23);
assert_eq!(foo.fetch_max(42, Ordering::SeqCst), 23);
assert_eq!(foo.load(Ordering::SeqCst), 42);

If you want to obtain the maximum value in one step, you can use the following:

use std::sync::atomic::{AtomicU8, Ordering};

let foo = AtomicU8::new(23);
let bar = 42;
let max_foo = foo.fetch_max(bar, Ordering::SeqCst).max(bar);
assert!(max_foo == 42);

pub fn fetch_min(&self, val: u8, order: Ordering) -> u8

Minimum with the current value.

Finds the minimum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_min takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u8.

Examples

use std::sync::atomic::{AtomicU8, Ordering};

let foo = AtomicU8::new(23);
assert_eq!(foo.fetch_min(42, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 23);
assert_eq!(foo.fetch_min(22, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 22);

If you want to obtain the minimum value in one step, you can use the following:

use std::sync::atomic::{AtomicU8, Ordering};

let foo = AtomicU8::new(23);
let bar = 12;
let min_foo = foo.fetch_min(bar, Ordering::SeqCst).min(bar);
assert_eq!(min_foo, 12);

pub const fn as_ptr(&self) -> *mut u8

Returns a mutable pointer to the underlying integer.

Doing non-atomic reads and writes on the resulting integer can be a data race. This method is mostly useful for FFI, where the function signature may use *mut u8 instead of &AtomicU8.

Returning an *mut pointer from a shared reference to this atomic is safe because the atomic types work with interior mutability. All modifications of an atomic change the value through a shared reference, and can do so safely as long as they use atomic operations. Any use of the returned raw pointer requires an unsafe block and still has to uphold the requirements of the memory model.

Examples

use std::sync::atomic::AtomicU8;

extern "C" {
    fn my_atomic_op(arg: *mut u8);
}

let atomic = AtomicU8::new(1);

// SAFETY: Safe as long as `my_atomic_op` is atomic.
unsafe {
    my_atomic_op(atomic.as_ptr());
}

impl Atomic<i16>

pub const fn new(v: i16) -> Self

Creates a new atomic integer.

Examples

use std::sync::atomic::AtomicI16;

let atomic_forty_two = AtomicI16::new(42);

pub const unsafe fn from_ptr<'a>(ptr: *mut i16) -> &'a AtomicI16

Creates a new reference to an atomic integer from a pointer.

Examples

use std::sync::atomic::{self, AtomicI16};

// Get a pointer to an allocated value
let ptr: *mut i16 = Box::into_raw(Box::new(0));

assert!(ptr.cast::<AtomicI16>().is_aligned());

{
    // Create an atomic view of the allocated value
    let atomic = unsafe {AtomicI16::from_ptr(ptr) };

    // Use `atomic` for atomic operations, possibly share it with other threads
    atomic.store(1, atomic::Ordering::Relaxed);
}

// It's ok to non-atomically access the value behind `ptr`,
// since the reference to the atomic ended its lifetime in the block above
assert_eq!(unsafe { *ptr }, 1);

// Deallocate the value
unsafe { drop(Box::from_raw(ptr)) }

Safety

• ptr must be aligned to align_of::<AtomicI16>() (note that on some platforms this can be bigger than align_of::<i16>()).
• ptr must be valid for both reads and writes for the whole lifetime 'a.
• You must adhere to the Memory model for atomic accesses. In particular, it is not allowed to mix conflicting atomic and non-atomic accesses, or atomic accesses of different sizes, without synchronization.

pub fn get_mut(&mut self) -> &mut i16

Returns a mutable reference to the underlying integer.

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let mut some_var = AtomicI16::new(10);
assert_eq!(*some_var.get_mut(), 10);
*some_var.get_mut() = 5;
assert_eq!(some_var.load(Ordering::SeqCst), 5);

pub fn from_mut(v: &mut i16) -> &mut Self

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut i16.

Note: This function is only available on targets where AtomicI16 has the same alignment as i16.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicI16, Ordering};

let mut some_int = 123;
let a = AtomicI16::from_mut(&mut some_int);
a.store(100, Ordering::Relaxed);
assert_eq!(some_int, 100);

pub fn get_mut_slice(this: &mut [Self]) -> &mut [i16]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get non-atomic access to a &mut [AtomicI16] slice

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicI16, Ordering};

let mut some_ints = [const { AtomicI16::new(0) }; 10];

let view: &mut [i16] = AtomicI16::get_mut_slice(&mut some_ints);
assert_eq!(view, [0; 10]);
view
    .iter_mut()
    .enumerate()
    .for_each(|(idx, int)| *int = idx as _);

std::thread::scope(|s| {
    some_ints
        .iter()
        .enumerate()
        .for_each(|(idx, int)| {
            s.spawn(move || assert_eq!(int.load(Ordering::Relaxed), idx as _));
        })
});

pub fn from_mut_slice(v: &mut [i16]) -> &mut [Self]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut [i16] slice.

Note: This function is only available on targets where AtomicI16 has the same alignment as i16.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicI16, Ordering};

let mut some_ints = [0; 10];
let a = &*AtomicI16::from_mut_slice(&mut some_ints);
std::thread::scope(|s| {
    for i in 0..a.len() {
        s.spawn(move || a[i].store(i as _, Ordering::Relaxed));
    }
});
for (i, n) in some_ints.into_iter().enumerate() {
    assert_eq!(i, n as usize);
}

pub const fn into_inner(self) -> i16

Consumes the atomic and returns the contained value.

This is safe because passing self by value guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::AtomicI16;

let some_var = AtomicI16::new(5);
assert_eq!(some_var.into_inner(), 5);

pub fn load(&self, order: Ordering) -> i16

Loads a value from the atomic integer.

load takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Acquire and Relaxed.

Panics

Panics if order is Release or AcqRel.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let some_var = AtomicI16::new(5);

assert_eq!(some_var.load(Ordering::Relaxed), 5);

pub fn store(&self, val: i16, order: Ordering)

Stores a value into the atomic integer.

store takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Release and Relaxed.

Panics

Panics if order is Acquire or AcqRel.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let some_var = AtomicI16::new(5);

some_var.store(10, Ordering::Relaxed);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn swap(&self, val: i16, order: Ordering) -> i16

Stores a value into the atomic integer, returning the previous value.

swap takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i16.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let some_var = AtomicI16::new(5);

assert_eq!(some_var.swap(10, Ordering::Relaxed), 5);

pub fn compare_and_swap(&self, current: i16, new: i16, order: Ordering) -> i16

👎Deprecated since 1.50.0: Use compare_exchange or compare_exchange_weak instead

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is always the previous value. If it is equal to current, then the value was updated.

compare_and_swap also takes an Ordering argument which describes the memory ordering of this operation. Notice that even when using AcqRel, the operation might fail and hence just perform an Acquire load, but not have Release semantics. Using Acquire makes the store part of this operation Relaxed if it happens, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i16.

Migrating to compare_exchange and compare_exchange_weak

compare_and_swap is equivalent to compare_exchange with the following mapping for memory orderings:

Original	Success	Failure
Relaxed	Relaxed	Relaxed
Acquire	Acquire	Acquire
Release	Release	Relaxed
AcqRel	AcqRel	Acquire
SeqCst	SeqCst	SeqCst

compare_and_swap and compare_exchange also differ in their return type. You can use compare_exchange(...).unwrap_or_else(|x| x) to recover the behavior of compare_and_swap, but in most cases it is more idiomatic to check whether the return value is Ok or Err rather than to infer success vs failure based on the value that was read.

During migration, consider whether it makes sense to use compare_exchange_weak instead. compare_exchange_weak is allowed to fail spuriously even when the comparison succeeds, which allows the compiler to generate better assembly code when the compare and swap is used in a loop.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let some_var = AtomicI16::new(5);

assert_eq!(some_var.compare_and_swap(5, 10, Ordering::Relaxed), 5);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_and_swap(6, 12, Ordering::Relaxed), 10);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn compare_exchange( &self, current: i16, new: i16, success: Ordering, failure: Ordering, ) -> Result<i16, i16>

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is a result indicating whether the new value was written and containing the previous value. On success this value is guaranteed to be equal to current.

compare_exchange takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on i16.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let some_var = AtomicI16::new(5);

assert_eq!(some_var.compare_exchange(5, 10,
                                     Ordering::Acquire,
                                     Ordering::Relaxed),
           Ok(5));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_exchange(6, 12,
                                     Ordering::SeqCst,
                                     Ordering::Acquire),
           Err(10));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim! This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn compare_exchange_weak( &self, current: i16, new: i16, success: Ordering, failure: Ordering, ) -> Result<i16, i16>

Stores a value into the atomic integer if the current value is the same as the current value.

Unlike AtomicI16::compare_exchange, this function is allowed to spuriously fail even when the comparison succeeds, which can result in more efficient code on some platforms. The return value is a result indicating whether the new value was written and containing the previous value.

compare_exchange_weak takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on i16.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let val = AtomicI16::new(4);

let mut old = val.load(Ordering::Relaxed);
loop {
    let new = old * 2;
    match val.compare_exchange_weak(old, new, Ordering::SeqCst, Ordering::Relaxed) {
        Ok(_) => break,
        Err(x) => old = x,
    }
}

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim. This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn fetch_add(&self, val: i16, order: Ordering) -> i16

Adds to the current value, returning the previous value.

This operation wraps around on overflow.

fetch_add takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i16.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let foo = AtomicI16::new(0);
assert_eq!(foo.fetch_add(10, Ordering::SeqCst), 0);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_sub(&self, val: i16, order: Ordering) -> i16

Subtracts from the current value, returning the previous value.

This operation wraps around on overflow.

fetch_sub takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i16.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let foo = AtomicI16::new(20);
assert_eq!(foo.fetch_sub(10, Ordering::SeqCst), 20);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_and(&self, val: i16, order: Ordering) -> i16

Bitwise “and” with the current value.

Performs a bitwise “and” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_and takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i16.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let foo = AtomicI16::new(0b101101);
assert_eq!(foo.fetch_and(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b100001);

pub fn fetch_nand(&self, val: i16, order: Ordering) -> i16

Bitwise “nand” with the current value.

Performs a bitwise “nand” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_nand takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i16.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let foo = AtomicI16::new(0x13);
assert_eq!(foo.fetch_nand(0x31, Ordering::SeqCst), 0x13);
assert_eq!(foo.load(Ordering::SeqCst), !(0x13 & 0x31));

pub fn fetch_or(&self, val: i16, order: Ordering) -> i16

Bitwise “or” with the current value.

Performs a bitwise “or” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_or takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i16.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let foo = AtomicI16::new(0b101101);
assert_eq!(foo.fetch_or(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b111111);

pub fn fetch_xor(&self, val: i16, order: Ordering) -> i16

Bitwise “xor” with the current value.

Performs a bitwise “xor” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_xor takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i16.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let foo = AtomicI16::new(0b101101);
assert_eq!(foo.fetch_xor(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b011110);

pub fn fetch_update<F>( &self, set_order: Ordering, fetch_order: Ordering, f: F, ) -> Result<i16, i16>

where F: FnMut(i16) -> Option<i16>,

👎Deprecating in 1.99.0: renamed to try_update for consistency

An alias for AtomicI16::try_update .

pub fn try_update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(i16) -> Option<i16>, ) -> Result<i16, i16>

Fetches the value, and applies a function to it that returns an optional new value. Returns a Result of Ok(previous_value) if the function returned Some(_), else Err(previous_value).

See also: update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, as long as the function returns Some(_), but the function will have been applied only once to the stored value.

try_update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicI16::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on i16.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let x = AtomicI16::new(7);
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |_| None), Err(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(8));
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(i16) -> i16, ) -> i16

Fetches the value, applies a function to it that it return a new value. The new value is stored and the old value is returned.

See also: try_update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, but the function will have been applied only once to the stored value.

update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicI16::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on i16.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let x = AtomicI16::new(7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 8);
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn fetch_max(&self, val: i16, order: Ordering) -> i16

Maximum with the current value.

Finds the maximum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_max takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i16.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let foo = AtomicI16::new(23);
assert_eq!(foo.fetch_max(42, Ordering::SeqCst), 23);
assert_eq!(foo.load(Ordering::SeqCst), 42);

If you want to obtain the maximum value in one step, you can use the following:

use std::sync::atomic::{AtomicI16, Ordering};

let foo = AtomicI16::new(23);
let bar = 42;
let max_foo = foo.fetch_max(bar, Ordering::SeqCst).max(bar);
assert!(max_foo == 42);

pub fn fetch_min(&self, val: i16, order: Ordering) -> i16

Minimum with the current value.

Finds the minimum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_min takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i16.

Examples

use std::sync::atomic::{AtomicI16, Ordering};

let foo = AtomicI16::new(23);
assert_eq!(foo.fetch_min(42, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 23);
assert_eq!(foo.fetch_min(22, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 22);

If you want to obtain the minimum value in one step, you can use the following:

use std::sync::atomic::{AtomicI16, Ordering};

let foo = AtomicI16::new(23);
let bar = 12;
let min_foo = foo.fetch_min(bar, Ordering::SeqCst).min(bar);
assert_eq!(min_foo, 12);

pub const fn as_ptr(&self) -> *mut i16

Returns a mutable pointer to the underlying integer.

Doing non-atomic reads and writes on the resulting integer can be a data race. This method is mostly useful for FFI, where the function signature may use *mut i16 instead of &AtomicI16.

Returning an *mut pointer from a shared reference to this atomic is safe because the atomic types work with interior mutability. All modifications of an atomic change the value through a shared reference, and can do so safely as long as they use atomic operations. Any use of the returned raw pointer requires an unsafe block and still has to uphold the requirements of the memory model.

Examples

use std::sync::atomic::AtomicI16;

extern "C" {
    fn my_atomic_op(arg: *mut i16);
}

let atomic = AtomicI16::new(1);

// SAFETY: Safe as long as `my_atomic_op` is atomic.
unsafe {
    my_atomic_op(atomic.as_ptr());
}

impl Atomic<u16>

pub const fn new(v: u16) -> Self

Creates a new atomic integer.

Examples

use std::sync::atomic::AtomicU16;

let atomic_forty_two = AtomicU16::new(42);

pub const unsafe fn from_ptr<'a>(ptr: *mut u16) -> &'a AtomicU16

Creates a new reference to an atomic integer from a pointer.

Examples

use std::sync::atomic::{self, AtomicU16};

// Get a pointer to an allocated value
let ptr: *mut u16 = Box::into_raw(Box::new(0));

assert!(ptr.cast::<AtomicU16>().is_aligned());

{
    // Create an atomic view of the allocated value
    let atomic = unsafe {AtomicU16::from_ptr(ptr) };

    // Use `atomic` for atomic operations, possibly share it with other threads
    atomic.store(1, atomic::Ordering::Relaxed);
}

// It's ok to non-atomically access the value behind `ptr`,
// since the reference to the atomic ended its lifetime in the block above
assert_eq!(unsafe { *ptr }, 1);

// Deallocate the value
unsafe { drop(Box::from_raw(ptr)) }

Safety

• ptr must be aligned to align_of::<AtomicU16>() (note that on some platforms this can be bigger than align_of::<u16>()).
• ptr must be valid for both reads and writes for the whole lifetime 'a.
• You must adhere to the Memory model for atomic accesses. In particular, it is not allowed to mix conflicting atomic and non-atomic accesses, or atomic accesses of different sizes, without synchronization.

pub fn get_mut(&mut self) -> &mut u16

Returns a mutable reference to the underlying integer.

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let mut some_var = AtomicU16::new(10);
assert_eq!(*some_var.get_mut(), 10);
*some_var.get_mut() = 5;
assert_eq!(some_var.load(Ordering::SeqCst), 5);

pub fn from_mut(v: &mut u16) -> &mut Self

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut u16.

Note: This function is only available on targets where AtomicU16 has the same alignment as u16.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicU16, Ordering};

let mut some_int = 123;
let a = AtomicU16::from_mut(&mut some_int);
a.store(100, Ordering::Relaxed);
assert_eq!(some_int, 100);

pub fn get_mut_slice(this: &mut [Self]) -> &mut [u16]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get non-atomic access to a &mut [AtomicU16] slice

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicU16, Ordering};

let mut some_ints = [const { AtomicU16::new(0) }; 10];

let view: &mut [u16] = AtomicU16::get_mut_slice(&mut some_ints);
assert_eq!(view, [0; 10]);
view
    .iter_mut()
    .enumerate()
    .for_each(|(idx, int)| *int = idx as _);

std::thread::scope(|s| {
    some_ints
        .iter()
        .enumerate()
        .for_each(|(idx, int)| {
            s.spawn(move || assert_eq!(int.load(Ordering::Relaxed), idx as _));
        })
});

pub fn from_mut_slice(v: &mut [u16]) -> &mut [Self]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut [u16] slice.

Note: This function is only available on targets where AtomicU16 has the same alignment as u16.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicU16, Ordering};

let mut some_ints = [0; 10];
let a = &*AtomicU16::from_mut_slice(&mut some_ints);
std::thread::scope(|s| {
    for i in 0..a.len() {
        s.spawn(move || a[i].store(i as _, Ordering::Relaxed));
    }
});
for (i, n) in some_ints.into_iter().enumerate() {
    assert_eq!(i, n as usize);
}

pub const fn into_inner(self) -> u16

Consumes the atomic and returns the contained value.

This is safe because passing self by value guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::AtomicU16;

let some_var = AtomicU16::new(5);
assert_eq!(some_var.into_inner(), 5);

pub fn load(&self, order: Ordering) -> u16

Loads a value from the atomic integer.

load takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Acquire and Relaxed.

Panics

Panics if order is Release or AcqRel.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let some_var = AtomicU16::new(5);

assert_eq!(some_var.load(Ordering::Relaxed), 5);

pub fn store(&self, val: u16, order: Ordering)

Stores a value into the atomic integer.

store takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Release and Relaxed.

Panics

Panics if order is Acquire or AcqRel.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let some_var = AtomicU16::new(5);

some_var.store(10, Ordering::Relaxed);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn swap(&self, val: u16, order: Ordering) -> u16

Stores a value into the atomic integer, returning the previous value.

swap takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u16.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let some_var = AtomicU16::new(5);

assert_eq!(some_var.swap(10, Ordering::Relaxed), 5);

pub fn compare_and_swap(&self, current: u16, new: u16, order: Ordering) -> u16

👎Deprecated since 1.50.0: Use compare_exchange or compare_exchange_weak instead

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is always the previous value. If it is equal to current, then the value was updated.

compare_and_swap also takes an Ordering argument which describes the memory ordering of this operation. Notice that even when using AcqRel, the operation might fail and hence just perform an Acquire load, but not have Release semantics. Using Acquire makes the store part of this operation Relaxed if it happens, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u16.

Migrating to compare_exchange and compare_exchange_weak

compare_and_swap is equivalent to compare_exchange with the following mapping for memory orderings:

Original	Success	Failure
Relaxed	Relaxed	Relaxed
Acquire	Acquire	Acquire
Release	Release	Relaxed
AcqRel	AcqRel	Acquire
SeqCst	SeqCst	SeqCst

compare_and_swap and compare_exchange also differ in their return type. You can use compare_exchange(...).unwrap_or_else(|x| x) to recover the behavior of compare_and_swap, but in most cases it is more idiomatic to check whether the return value is Ok or Err rather than to infer success vs failure based on the value that was read.

During migration, consider whether it makes sense to use compare_exchange_weak instead. compare_exchange_weak is allowed to fail spuriously even when the comparison succeeds, which allows the compiler to generate better assembly code when the compare and swap is used in a loop.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let some_var = AtomicU16::new(5);

assert_eq!(some_var.compare_and_swap(5, 10, Ordering::Relaxed), 5);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_and_swap(6, 12, Ordering::Relaxed), 10);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn compare_exchange( &self, current: u16, new: u16, success: Ordering, failure: Ordering, ) -> Result<u16, u16>

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is a result indicating whether the new value was written and containing the previous value. On success this value is guaranteed to be equal to current.

compare_exchange takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u16.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let some_var = AtomicU16::new(5);

assert_eq!(some_var.compare_exchange(5, 10,
                                     Ordering::Acquire,
                                     Ordering::Relaxed),
           Ok(5));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_exchange(6, 12,
                                     Ordering::SeqCst,
                                     Ordering::Acquire),
           Err(10));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim! This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn compare_exchange_weak( &self, current: u16, new: u16, success: Ordering, failure: Ordering, ) -> Result<u16, u16>

Stores a value into the atomic integer if the current value is the same as the current value.

Unlike AtomicU16::compare_exchange, this function is allowed to spuriously fail even when the comparison succeeds, which can result in more efficient code on some platforms. The return value is a result indicating whether the new value was written and containing the previous value.

compare_exchange_weak takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u16.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let val = AtomicU16::new(4);

let mut old = val.load(Ordering::Relaxed);
loop {
    let new = old * 2;
    match val.compare_exchange_weak(old, new, Ordering::SeqCst, Ordering::Relaxed) {
        Ok(_) => break,
        Err(x) => old = x,
    }
}

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim. This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn fetch_add(&self, val: u16, order: Ordering) -> u16

Adds to the current value, returning the previous value.

This operation wraps around on overflow.

fetch_add takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u16.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let foo = AtomicU16::new(0);
assert_eq!(foo.fetch_add(10, Ordering::SeqCst), 0);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_sub(&self, val: u16, order: Ordering) -> u16

Subtracts from the current value, returning the previous value.

This operation wraps around on overflow.

fetch_sub takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u16.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let foo = AtomicU16::new(20);
assert_eq!(foo.fetch_sub(10, Ordering::SeqCst), 20);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_and(&self, val: u16, order: Ordering) -> u16

Bitwise “and” with the current value.

Performs a bitwise “and” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_and takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u16.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let foo = AtomicU16::new(0b101101);
assert_eq!(foo.fetch_and(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b100001);

pub fn fetch_nand(&self, val: u16, order: Ordering) -> u16

Bitwise “nand” with the current value.

Performs a bitwise “nand” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_nand takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u16.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let foo = AtomicU16::new(0x13);
assert_eq!(foo.fetch_nand(0x31, Ordering::SeqCst), 0x13);
assert_eq!(foo.load(Ordering::SeqCst), !(0x13 & 0x31));

pub fn fetch_or(&self, val: u16, order: Ordering) -> u16

Bitwise “or” with the current value.

Performs a bitwise “or” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_or takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u16.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let foo = AtomicU16::new(0b101101);
assert_eq!(foo.fetch_or(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b111111);

pub fn fetch_xor(&self, val: u16, order: Ordering) -> u16

Bitwise “xor” with the current value.

Performs a bitwise “xor” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_xor takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u16.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let foo = AtomicU16::new(0b101101);
assert_eq!(foo.fetch_xor(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b011110);

pub fn fetch_update<F>( &self, set_order: Ordering, fetch_order: Ordering, f: F, ) -> Result<u16, u16>

where F: FnMut(u16) -> Option<u16>,

👎Deprecating in 1.99.0: renamed to try_update for consistency

An alias for AtomicU16::try_update .

pub fn try_update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(u16) -> Option<u16>, ) -> Result<u16, u16>

Fetches the value, and applies a function to it that returns an optional new value. Returns a Result of Ok(previous_value) if the function returned Some(_), else Err(previous_value).

See also: update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, as long as the function returns Some(_), but the function will have been applied only once to the stored value.

try_update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicU16::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u16.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let x = AtomicU16::new(7);
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |_| None), Err(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(8));
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(u16) -> u16, ) -> u16

Fetches the value, applies a function to it that it return a new value. The new value is stored and the old value is returned.

See also: try_update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, but the function will have been applied only once to the stored value.

update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicU16::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u16.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let x = AtomicU16::new(7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 8);
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn fetch_max(&self, val: u16, order: Ordering) -> u16

Maximum with the current value.

Finds the maximum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_max takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u16.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let foo = AtomicU16::new(23);
assert_eq!(foo.fetch_max(42, Ordering::SeqCst), 23);
assert_eq!(foo.load(Ordering::SeqCst), 42);

If you want to obtain the maximum value in one step, you can use the following:

use std::sync::atomic::{AtomicU16, Ordering};

let foo = AtomicU16::new(23);
let bar = 42;
let max_foo = foo.fetch_max(bar, Ordering::SeqCst).max(bar);
assert!(max_foo == 42);

pub fn fetch_min(&self, val: u16, order: Ordering) -> u16

Minimum with the current value.

Finds the minimum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_min takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u16.

Examples

use std::sync::atomic::{AtomicU16, Ordering};

let foo = AtomicU16::new(23);
assert_eq!(foo.fetch_min(42, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 23);
assert_eq!(foo.fetch_min(22, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 22);

If you want to obtain the minimum value in one step, you can use the following:

use std::sync::atomic::{AtomicU16, Ordering};

let foo = AtomicU16::new(23);
let bar = 12;
let min_foo = foo.fetch_min(bar, Ordering::SeqCst).min(bar);
assert_eq!(min_foo, 12);

pub const fn as_ptr(&self) -> *mut u16

Returns a mutable pointer to the underlying integer.

Doing non-atomic reads and writes on the resulting integer can be a data race. This method is mostly useful for FFI, where the function signature may use *mut u16 instead of &AtomicU16.

Returning an *mut pointer from a shared reference to this atomic is safe because the atomic types work with interior mutability. All modifications of an atomic change the value through a shared reference, and can do so safely as long as they use atomic operations. Any use of the returned raw pointer requires an unsafe block and still has to uphold the requirements of the memory model.

Examples

use std::sync::atomic::AtomicU16;

extern "C" {
    fn my_atomic_op(arg: *mut u16);
}

let atomic = AtomicU16::new(1);

// SAFETY: Safe as long as `my_atomic_op` is atomic.
unsafe {
    my_atomic_op(atomic.as_ptr());
}

impl Atomic<i32>

pub const fn new(v: i32) -> Self

Creates a new atomic integer.

Examples

use std::sync::atomic::AtomicI32;

let atomic_forty_two = AtomicI32::new(42);

pub const unsafe fn from_ptr<'a>(ptr: *mut i32) -> &'a AtomicI32

Creates a new reference to an atomic integer from a pointer.

Examples

use std::sync::atomic::{self, AtomicI32};

// Get a pointer to an allocated value
let ptr: *mut i32 = Box::into_raw(Box::new(0));

assert!(ptr.cast::<AtomicI32>().is_aligned());

{
    // Create an atomic view of the allocated value
    let atomic = unsafe {AtomicI32::from_ptr(ptr) };

    // Use `atomic` for atomic operations, possibly share it with other threads
    atomic.store(1, atomic::Ordering::Relaxed);
}

// It's ok to non-atomically access the value behind `ptr`,
// since the reference to the atomic ended its lifetime in the block above
assert_eq!(unsafe { *ptr }, 1);

// Deallocate the value
unsafe { drop(Box::from_raw(ptr)) }

Safety

• ptr must be aligned to align_of::<AtomicI32>() (note that on some platforms this can be bigger than align_of::<i32>()).
• ptr must be valid for both reads and writes for the whole lifetime 'a.
• You must adhere to the Memory model for atomic accesses. In particular, it is not allowed to mix conflicting atomic and non-atomic accesses, or atomic accesses of different sizes, without synchronization.

pub fn get_mut(&mut self) -> &mut i32

Returns a mutable reference to the underlying integer.

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let mut some_var = AtomicI32::new(10);
assert_eq!(*some_var.get_mut(), 10);
*some_var.get_mut() = 5;
assert_eq!(some_var.load(Ordering::SeqCst), 5);

pub fn from_mut(v: &mut i32) -> &mut Self

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut i32.

Note: This function is only available on targets where AtomicI32 has the same alignment as i32.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicI32, Ordering};

let mut some_int = 123;
let a = AtomicI32::from_mut(&mut some_int);
a.store(100, Ordering::Relaxed);
assert_eq!(some_int, 100);

pub fn get_mut_slice(this: &mut [Self]) -> &mut [i32]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get non-atomic access to a &mut [AtomicI32] slice

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicI32, Ordering};

let mut some_ints = [const { AtomicI32::new(0) }; 10];

let view: &mut [i32] = AtomicI32::get_mut_slice(&mut some_ints);
assert_eq!(view, [0; 10]);
view
    .iter_mut()
    .enumerate()
    .for_each(|(idx, int)| *int = idx as _);

std::thread::scope(|s| {
    some_ints
        .iter()
        .enumerate()
        .for_each(|(idx, int)| {
            s.spawn(move || assert_eq!(int.load(Ordering::Relaxed), idx as _));
        })
});

pub fn from_mut_slice(v: &mut [i32]) -> &mut [Self]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut [i32] slice.

Note: This function is only available on targets where AtomicI32 has the same alignment as i32.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicI32, Ordering};

let mut some_ints = [0; 10];
let a = &*AtomicI32::from_mut_slice(&mut some_ints);
std::thread::scope(|s| {
    for i in 0..a.len() {
        s.spawn(move || a[i].store(i as _, Ordering::Relaxed));
    }
});
for (i, n) in some_ints.into_iter().enumerate() {
    assert_eq!(i, n as usize);
}

pub const fn into_inner(self) -> i32

Consumes the atomic and returns the contained value.

This is safe because passing self by value guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::AtomicI32;

let some_var = AtomicI32::new(5);
assert_eq!(some_var.into_inner(), 5);

pub fn load(&self, order: Ordering) -> i32

Loads a value from the atomic integer.

load takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Acquire and Relaxed.

Panics

Panics if order is Release or AcqRel.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let some_var = AtomicI32::new(5);

assert_eq!(some_var.load(Ordering::Relaxed), 5);

pub fn store(&self, val: i32, order: Ordering)

Stores a value into the atomic integer.

store takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Release and Relaxed.

Panics

Panics if order is Acquire or AcqRel.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let some_var = AtomicI32::new(5);

some_var.store(10, Ordering::Relaxed);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn swap(&self, val: i32, order: Ordering) -> i32

Stores a value into the atomic integer, returning the previous value.

swap takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i32.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let some_var = AtomicI32::new(5);

assert_eq!(some_var.swap(10, Ordering::Relaxed), 5);

pub fn compare_and_swap(&self, current: i32, new: i32, order: Ordering) -> i32

👎Deprecated since 1.50.0: Use compare_exchange or compare_exchange_weak instead

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is always the previous value. If it is equal to current, then the value was updated.

compare_and_swap also takes an Ordering argument which describes the memory ordering of this operation. Notice that even when using AcqRel, the operation might fail and hence just perform an Acquire load, but not have Release semantics. Using Acquire makes the store part of this operation Relaxed if it happens, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i32.

Migrating to compare_exchange and compare_exchange_weak

compare_and_swap is equivalent to compare_exchange with the following mapping for memory orderings:

Original	Success	Failure
Relaxed	Relaxed	Relaxed
Acquire	Acquire	Acquire
Release	Release	Relaxed
AcqRel	AcqRel	Acquire
SeqCst	SeqCst	SeqCst

compare_and_swap and compare_exchange also differ in their return type. You can use compare_exchange(...).unwrap_or_else(|x| x) to recover the behavior of compare_and_swap, but in most cases it is more idiomatic to check whether the return value is Ok or Err rather than to infer success vs failure based on the value that was read.

During migration, consider whether it makes sense to use compare_exchange_weak instead. compare_exchange_weak is allowed to fail spuriously even when the comparison succeeds, which allows the compiler to generate better assembly code when the compare and swap is used in a loop.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let some_var = AtomicI32::new(5);

assert_eq!(some_var.compare_and_swap(5, 10, Ordering::Relaxed), 5);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_and_swap(6, 12, Ordering::Relaxed), 10);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn compare_exchange( &self, current: i32, new: i32, success: Ordering, failure: Ordering, ) -> Result<i32, i32>

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is a result indicating whether the new value was written and containing the previous value. On success this value is guaranteed to be equal to current.

compare_exchange takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on i32.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let some_var = AtomicI32::new(5);

assert_eq!(some_var.compare_exchange(5, 10,
                                     Ordering::Acquire,
                                     Ordering::Relaxed),
           Ok(5));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_exchange(6, 12,
                                     Ordering::SeqCst,
                                     Ordering::Acquire),
           Err(10));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim! This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn compare_exchange_weak( &self, current: i32, new: i32, success: Ordering, failure: Ordering, ) -> Result<i32, i32>

Stores a value into the atomic integer if the current value is the same as the current value.

Unlike AtomicI32::compare_exchange, this function is allowed to spuriously fail even when the comparison succeeds, which can result in more efficient code on some platforms. The return value is a result indicating whether the new value was written and containing the previous value.

compare_exchange_weak takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on i32.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let val = AtomicI32::new(4);

let mut old = val.load(Ordering::Relaxed);
loop {
    let new = old * 2;
    match val.compare_exchange_weak(old, new, Ordering::SeqCst, Ordering::Relaxed) {
        Ok(_) => break,
        Err(x) => old = x,
    }
}

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim. This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn fetch_add(&self, val: i32, order: Ordering) -> i32

Adds to the current value, returning the previous value.

This operation wraps around on overflow.

fetch_add takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i32.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let foo = AtomicI32::new(0);
assert_eq!(foo.fetch_add(10, Ordering::SeqCst), 0);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_sub(&self, val: i32, order: Ordering) -> i32

Subtracts from the current value, returning the previous value.

This operation wraps around on overflow.

fetch_sub takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i32.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let foo = AtomicI32::new(20);
assert_eq!(foo.fetch_sub(10, Ordering::SeqCst), 20);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_and(&self, val: i32, order: Ordering) -> i32

Bitwise “and” with the current value.

Performs a bitwise “and” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_and takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i32.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let foo = AtomicI32::new(0b101101);
assert_eq!(foo.fetch_and(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b100001);

pub fn fetch_nand(&self, val: i32, order: Ordering) -> i32

Bitwise “nand” with the current value.

Performs a bitwise “nand” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_nand takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i32.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let foo = AtomicI32::new(0x13);
assert_eq!(foo.fetch_nand(0x31, Ordering::SeqCst), 0x13);
assert_eq!(foo.load(Ordering::SeqCst), !(0x13 & 0x31));

pub fn fetch_or(&self, val: i32, order: Ordering) -> i32

Bitwise “or” with the current value.

Performs a bitwise “or” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_or takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i32.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let foo = AtomicI32::new(0b101101);
assert_eq!(foo.fetch_or(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b111111);

pub fn fetch_xor(&self, val: i32, order: Ordering) -> i32

Bitwise “xor” with the current value.

Performs a bitwise “xor” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_xor takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i32.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let foo = AtomicI32::new(0b101101);
assert_eq!(foo.fetch_xor(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b011110);

pub fn fetch_update<F>( &self, set_order: Ordering, fetch_order: Ordering, f: F, ) -> Result<i32, i32>

where F: FnMut(i32) -> Option<i32>,

👎Deprecating in 1.99.0: renamed to try_update for consistency

An alias for AtomicI32::try_update .

pub fn try_update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(i32) -> Option<i32>, ) -> Result<i32, i32>

Fetches the value, and applies a function to it that returns an optional new value. Returns a Result of Ok(previous_value) if the function returned Some(_), else Err(previous_value).

See also: update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, as long as the function returns Some(_), but the function will have been applied only once to the stored value.

try_update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicI32::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on i32.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let x = AtomicI32::new(7);
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |_| None), Err(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(8));
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(i32) -> i32, ) -> i32

Fetches the value, applies a function to it that it return a new value. The new value is stored and the old value is returned.

See also: try_update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, but the function will have been applied only once to the stored value.

update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicI32::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on i32.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let x = AtomicI32::new(7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 8);
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn fetch_max(&self, val: i32, order: Ordering) -> i32

Maximum with the current value.

Finds the maximum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_max takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i32.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let foo = AtomicI32::new(23);
assert_eq!(foo.fetch_max(42, Ordering::SeqCst), 23);
assert_eq!(foo.load(Ordering::SeqCst), 42);

If you want to obtain the maximum value in one step, you can use the following:

use std::sync::atomic::{AtomicI32, Ordering};

let foo = AtomicI32::new(23);
let bar = 42;
let max_foo = foo.fetch_max(bar, Ordering::SeqCst).max(bar);
assert!(max_foo == 42);

pub fn fetch_min(&self, val: i32, order: Ordering) -> i32

Minimum with the current value.

Finds the minimum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_min takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i32.

Examples

use std::sync::atomic::{AtomicI32, Ordering};

let foo = AtomicI32::new(23);
assert_eq!(foo.fetch_min(42, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 23);
assert_eq!(foo.fetch_min(22, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 22);

If you want to obtain the minimum value in one step, you can use the following:

use std::sync::atomic::{AtomicI32, Ordering};

let foo = AtomicI32::new(23);
let bar = 12;
let min_foo = foo.fetch_min(bar, Ordering::SeqCst).min(bar);
assert_eq!(min_foo, 12);

pub const fn as_ptr(&self) -> *mut i32

Returns a mutable pointer to the underlying integer.

Doing non-atomic reads and writes on the resulting integer can be a data race. This method is mostly useful for FFI, where the function signature may use *mut i32 instead of &AtomicI32.

Returning an *mut pointer from a shared reference to this atomic is safe because the atomic types work with interior mutability. All modifications of an atomic change the value through a shared reference, and can do so safely as long as they use atomic operations. Any use of the returned raw pointer requires an unsafe block and still has to uphold the requirements of the memory model.

Examples

use std::sync::atomic::AtomicI32;

extern "C" {
    fn my_atomic_op(arg: *mut i32);
}

let atomic = AtomicI32::new(1);

// SAFETY: Safe as long as `my_atomic_op` is atomic.
unsafe {
    my_atomic_op(atomic.as_ptr());
}

impl Atomic<u32>

pub const fn new(v: u32) -> Self

Creates a new atomic integer.

Examples

use std::sync::atomic::AtomicU32;

let atomic_forty_two = AtomicU32::new(42);

pub const unsafe fn from_ptr<'a>(ptr: *mut u32) -> &'a AtomicU32

Creates a new reference to an atomic integer from a pointer.

Examples

use std::sync::atomic::{self, AtomicU32};

// Get a pointer to an allocated value
let ptr: *mut u32 = Box::into_raw(Box::new(0));

assert!(ptr.cast::<AtomicU32>().is_aligned());

{
    // Create an atomic view of the allocated value
    let atomic = unsafe {AtomicU32::from_ptr(ptr) };

    // Use `atomic` for atomic operations, possibly share it with other threads
    atomic.store(1, atomic::Ordering::Relaxed);
}

// It's ok to non-atomically access the value behind `ptr`,
// since the reference to the atomic ended its lifetime in the block above
assert_eq!(unsafe { *ptr }, 1);

// Deallocate the value
unsafe { drop(Box::from_raw(ptr)) }

Safety

• ptr must be aligned to align_of::<AtomicU32>() (note that on some platforms this can be bigger than align_of::<u32>()).
• ptr must be valid for both reads and writes for the whole lifetime 'a.
• You must adhere to the Memory model for atomic accesses. In particular, it is not allowed to mix conflicting atomic and non-atomic accesses, or atomic accesses of different sizes, without synchronization.

pub fn get_mut(&mut self) -> &mut u32

Returns a mutable reference to the underlying integer.

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let mut some_var = AtomicU32::new(10);
assert_eq!(*some_var.get_mut(), 10);
*some_var.get_mut() = 5;
assert_eq!(some_var.load(Ordering::SeqCst), 5);

pub fn from_mut(v: &mut u32) -> &mut Self

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut u32.

Note: This function is only available on targets where AtomicU32 has the same alignment as u32.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicU32, Ordering};

let mut some_int = 123;
let a = AtomicU32::from_mut(&mut some_int);
a.store(100, Ordering::Relaxed);
assert_eq!(some_int, 100);

pub fn get_mut_slice(this: &mut [Self]) -> &mut [u32]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get non-atomic access to a &mut [AtomicU32] slice

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicU32, Ordering};

let mut some_ints = [const { AtomicU32::new(0) }; 10];

let view: &mut [u32] = AtomicU32::get_mut_slice(&mut some_ints);
assert_eq!(view, [0; 10]);
view
    .iter_mut()
    .enumerate()
    .for_each(|(idx, int)| *int = idx as _);

std::thread::scope(|s| {
    some_ints
        .iter()
        .enumerate()
        .for_each(|(idx, int)| {
            s.spawn(move || assert_eq!(int.load(Ordering::Relaxed), idx as _));
        })
});

pub fn from_mut_slice(v: &mut [u32]) -> &mut [Self]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut [u32] slice.

Note: This function is only available on targets where AtomicU32 has the same alignment as u32.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicU32, Ordering};

let mut some_ints = [0; 10];
let a = &*AtomicU32::from_mut_slice(&mut some_ints);
std::thread::scope(|s| {
    for i in 0..a.len() {
        s.spawn(move || a[i].store(i as _, Ordering::Relaxed));
    }
});
for (i, n) in some_ints.into_iter().enumerate() {
    assert_eq!(i, n as usize);
}

pub const fn into_inner(self) -> u32

Consumes the atomic and returns the contained value.

This is safe because passing self by value guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::AtomicU32;

let some_var = AtomicU32::new(5);
assert_eq!(some_var.into_inner(), 5);

pub fn load(&self, order: Ordering) -> u32

Loads a value from the atomic integer.

load takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Acquire and Relaxed.

Panics

Panics if order is Release or AcqRel.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let some_var = AtomicU32::new(5);

assert_eq!(some_var.load(Ordering::Relaxed), 5);

pub fn store(&self, val: u32, order: Ordering)

Stores a value into the atomic integer.

store takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Release and Relaxed.

Panics

Panics if order is Acquire or AcqRel.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let some_var = AtomicU32::new(5);

some_var.store(10, Ordering::Relaxed);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn swap(&self, val: u32, order: Ordering) -> u32

Stores a value into the atomic integer, returning the previous value.

swap takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u32.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let some_var = AtomicU32::new(5);

assert_eq!(some_var.swap(10, Ordering::Relaxed), 5);

pub fn compare_and_swap(&self, current: u32, new: u32, order: Ordering) -> u32

👎Deprecated since 1.50.0: Use compare_exchange or compare_exchange_weak instead

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is always the previous value. If it is equal to current, then the value was updated.

compare_and_swap also takes an Ordering argument which describes the memory ordering of this operation. Notice that even when using AcqRel, the operation might fail and hence just perform an Acquire load, but not have Release semantics. Using Acquire makes the store part of this operation Relaxed if it happens, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u32.

Migrating to compare_exchange and compare_exchange_weak

compare_and_swap is equivalent to compare_exchange with the following mapping for memory orderings:

Original	Success	Failure
Relaxed	Relaxed	Relaxed
Acquire	Acquire	Acquire
Release	Release	Relaxed
AcqRel	AcqRel	Acquire
SeqCst	SeqCst	SeqCst

compare_and_swap and compare_exchange also differ in their return type. You can use compare_exchange(...).unwrap_or_else(|x| x) to recover the behavior of compare_and_swap, but in most cases it is more idiomatic to check whether the return value is Ok or Err rather than to infer success vs failure based on the value that was read.

During migration, consider whether it makes sense to use compare_exchange_weak instead. compare_exchange_weak is allowed to fail spuriously even when the comparison succeeds, which allows the compiler to generate better assembly code when the compare and swap is used in a loop.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let some_var = AtomicU32::new(5);

assert_eq!(some_var.compare_and_swap(5, 10, Ordering::Relaxed), 5);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_and_swap(6, 12, Ordering::Relaxed), 10);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn compare_exchange( &self, current: u32, new: u32, success: Ordering, failure: Ordering, ) -> Result<u32, u32>

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is a result indicating whether the new value was written and containing the previous value. On success this value is guaranteed to be equal to current.

compare_exchange takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u32.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let some_var = AtomicU32::new(5);

assert_eq!(some_var.compare_exchange(5, 10,
                                     Ordering::Acquire,
                                     Ordering::Relaxed),
           Ok(5));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_exchange(6, 12,
                                     Ordering::SeqCst,
                                     Ordering::Acquire),
           Err(10));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim! This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn compare_exchange_weak( &self, current: u32, new: u32, success: Ordering, failure: Ordering, ) -> Result<u32, u32>

Stores a value into the atomic integer if the current value is the same as the current value.

Unlike AtomicU32::compare_exchange, this function is allowed to spuriously fail even when the comparison succeeds, which can result in more efficient code on some platforms. The return value is a result indicating whether the new value was written and containing the previous value.

compare_exchange_weak takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u32.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let val = AtomicU32::new(4);

let mut old = val.load(Ordering::Relaxed);
loop {
    let new = old * 2;
    match val.compare_exchange_weak(old, new, Ordering::SeqCst, Ordering::Relaxed) {
        Ok(_) => break,
        Err(x) => old = x,
    }
}

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim. This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn fetch_add(&self, val: u32, order: Ordering) -> u32

Adds to the current value, returning the previous value.

This operation wraps around on overflow.

fetch_add takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u32.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let foo = AtomicU32::new(0);
assert_eq!(foo.fetch_add(10, Ordering::SeqCst), 0);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_sub(&self, val: u32, order: Ordering) -> u32

Subtracts from the current value, returning the previous value.

This operation wraps around on overflow.

fetch_sub takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u32.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let foo = AtomicU32::new(20);
assert_eq!(foo.fetch_sub(10, Ordering::SeqCst), 20);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_and(&self, val: u32, order: Ordering) -> u32

Bitwise “and” with the current value.

Performs a bitwise “and” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_and takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u32.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let foo = AtomicU32::new(0b101101);
assert_eq!(foo.fetch_and(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b100001);

pub fn fetch_nand(&self, val: u32, order: Ordering) -> u32

Bitwise “nand” with the current value.

Performs a bitwise “nand” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_nand takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u32.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let foo = AtomicU32::new(0x13);
assert_eq!(foo.fetch_nand(0x31, Ordering::SeqCst), 0x13);
assert_eq!(foo.load(Ordering::SeqCst), !(0x13 & 0x31));

pub fn fetch_or(&self, val: u32, order: Ordering) -> u32

Bitwise “or” with the current value.

Performs a bitwise “or” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_or takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u32.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let foo = AtomicU32::new(0b101101);
assert_eq!(foo.fetch_or(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b111111);

pub fn fetch_xor(&self, val: u32, order: Ordering) -> u32

Bitwise “xor” with the current value.

Performs a bitwise “xor” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_xor takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u32.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let foo = AtomicU32::new(0b101101);
assert_eq!(foo.fetch_xor(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b011110);

pub fn fetch_update<F>( &self, set_order: Ordering, fetch_order: Ordering, f: F, ) -> Result<u32, u32>

where F: FnMut(u32) -> Option<u32>,

👎Deprecating in 1.99.0: renamed to try_update for consistency

An alias for AtomicU32::try_update .

pub fn try_update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(u32) -> Option<u32>, ) -> Result<u32, u32>

Fetches the value, and applies a function to it that returns an optional new value. Returns a Result of Ok(previous_value) if the function returned Some(_), else Err(previous_value).

See also: update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, as long as the function returns Some(_), but the function will have been applied only once to the stored value.

try_update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicU32::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u32.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let x = AtomicU32::new(7);
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |_| None), Err(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(8));
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(u32) -> u32, ) -> u32

Fetches the value, applies a function to it that it return a new value. The new value is stored and the old value is returned.

See also: try_update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, but the function will have been applied only once to the stored value.

update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicU32::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u32.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let x = AtomicU32::new(7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 8);
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn fetch_max(&self, val: u32, order: Ordering) -> u32

Maximum with the current value.

Finds the maximum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_max takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u32.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let foo = AtomicU32::new(23);
assert_eq!(foo.fetch_max(42, Ordering::SeqCst), 23);
assert_eq!(foo.load(Ordering::SeqCst), 42);

If you want to obtain the maximum value in one step, you can use the following:

use std::sync::atomic::{AtomicU32, Ordering};

let foo = AtomicU32::new(23);
let bar = 42;
let max_foo = foo.fetch_max(bar, Ordering::SeqCst).max(bar);
assert!(max_foo == 42);

pub fn fetch_min(&self, val: u32, order: Ordering) -> u32

Minimum with the current value.

Finds the minimum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_min takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u32.

Examples

use std::sync::atomic::{AtomicU32, Ordering};

let foo = AtomicU32::new(23);
assert_eq!(foo.fetch_min(42, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 23);
assert_eq!(foo.fetch_min(22, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 22);

If you want to obtain the minimum value in one step, you can use the following:

use std::sync::atomic::{AtomicU32, Ordering};

let foo = AtomicU32::new(23);
let bar = 12;
let min_foo = foo.fetch_min(bar, Ordering::SeqCst).min(bar);
assert_eq!(min_foo, 12);

pub const fn as_ptr(&self) -> *mut u32

Returns a mutable pointer to the underlying integer.

Doing non-atomic reads and writes on the resulting integer can be a data race. This method is mostly useful for FFI, where the function signature may use *mut u32 instead of &AtomicU32.

Returning an *mut pointer from a shared reference to this atomic is safe because the atomic types work with interior mutability. All modifications of an atomic change the value through a shared reference, and can do so safely as long as they use atomic operations. Any use of the returned raw pointer requires an unsafe block and still has to uphold the requirements of the memory model.

Examples

use std::sync::atomic::AtomicU32;

extern "C" {
    fn my_atomic_op(arg: *mut u32);
}

let atomic = AtomicU32::new(1);

// SAFETY: Safe as long as `my_atomic_op` is atomic.
unsafe {
    my_atomic_op(atomic.as_ptr());
}

impl Atomic<i64>

pub const fn new(v: i64) -> Self

Creates a new atomic integer.

Examples

use std::sync::atomic::AtomicI64;

let atomic_forty_two = AtomicI64::new(42);

pub const unsafe fn from_ptr<'a>(ptr: *mut i64) -> &'a AtomicI64

Creates a new reference to an atomic integer from a pointer.

Examples

use std::sync::atomic::{self, AtomicI64};

// Get a pointer to an allocated value
let ptr: *mut i64 = Box::into_raw(Box::new(0));

assert!(ptr.cast::<AtomicI64>().is_aligned());

{
    // Create an atomic view of the allocated value
    let atomic = unsafe {AtomicI64::from_ptr(ptr) };

    // Use `atomic` for atomic operations, possibly share it with other threads
    atomic.store(1, atomic::Ordering::Relaxed);
}

// It's ok to non-atomically access the value behind `ptr`,
// since the reference to the atomic ended its lifetime in the block above
assert_eq!(unsafe { *ptr }, 1);

// Deallocate the value
unsafe { drop(Box::from_raw(ptr)) }

Safety

• ptr must be aligned to align_of::<AtomicI64>() (note that on some platforms this can be bigger than align_of::<i64>()).
• ptr must be valid for both reads and writes for the whole lifetime 'a.
• You must adhere to the Memory model for atomic accesses. In particular, it is not allowed to mix conflicting atomic and non-atomic accesses, or atomic accesses of different sizes, without synchronization.

pub fn get_mut(&mut self) -> &mut i64

Returns a mutable reference to the underlying integer.

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let mut some_var = AtomicI64::new(10);
assert_eq!(*some_var.get_mut(), 10);
*some_var.get_mut() = 5;
assert_eq!(some_var.load(Ordering::SeqCst), 5);

pub fn from_mut(v: &mut i64) -> &mut Self

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut i64.

Note: This function is only available on targets where AtomicI64 has the same alignment as i64.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicI64, Ordering};

let mut some_int = 123;
let a = AtomicI64::from_mut(&mut some_int);
a.store(100, Ordering::Relaxed);
assert_eq!(some_int, 100);

pub fn get_mut_slice(this: &mut [Self]) -> &mut [i64]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get non-atomic access to a &mut [AtomicI64] slice

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicI64, Ordering};

let mut some_ints = [const { AtomicI64::new(0) }; 10];

let view: &mut [i64] = AtomicI64::get_mut_slice(&mut some_ints);
assert_eq!(view, [0; 10]);
view
    .iter_mut()
    .enumerate()
    .for_each(|(idx, int)| *int = idx as _);

std::thread::scope(|s| {
    some_ints
        .iter()
        .enumerate()
        .for_each(|(idx, int)| {
            s.spawn(move || assert_eq!(int.load(Ordering::Relaxed), idx as _));
        })
});

pub fn from_mut_slice(v: &mut [i64]) -> &mut [Self]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut [i64] slice.

Note: This function is only available on targets where AtomicI64 has the same alignment as i64.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicI64, Ordering};

let mut some_ints = [0; 10];
let a = &*AtomicI64::from_mut_slice(&mut some_ints);
std::thread::scope(|s| {
    for i in 0..a.len() {
        s.spawn(move || a[i].store(i as _, Ordering::Relaxed));
    }
});
for (i, n) in some_ints.into_iter().enumerate() {
    assert_eq!(i, n as usize);
}

pub const fn into_inner(self) -> i64

Consumes the atomic and returns the contained value.

This is safe because passing self by value guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::AtomicI64;

let some_var = AtomicI64::new(5);
assert_eq!(some_var.into_inner(), 5);

pub fn load(&self, order: Ordering) -> i64

Loads a value from the atomic integer.

load takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Acquire and Relaxed.

Panics

Panics if order is Release or AcqRel.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let some_var = AtomicI64::new(5);

assert_eq!(some_var.load(Ordering::Relaxed), 5);

pub fn store(&self, val: i64, order: Ordering)

Stores a value into the atomic integer.

store takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Release and Relaxed.

Panics

Panics if order is Acquire or AcqRel.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let some_var = AtomicI64::new(5);

some_var.store(10, Ordering::Relaxed);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn swap(&self, val: i64, order: Ordering) -> i64

Stores a value into the atomic integer, returning the previous value.

swap takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i64.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let some_var = AtomicI64::new(5);

assert_eq!(some_var.swap(10, Ordering::Relaxed), 5);

pub fn compare_and_swap(&self, current: i64, new: i64, order: Ordering) -> i64

👎Deprecated since 1.50.0: Use compare_exchange or compare_exchange_weak instead

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is always the previous value. If it is equal to current, then the value was updated.

compare_and_swap also takes an Ordering argument which describes the memory ordering of this operation. Notice that even when using AcqRel, the operation might fail and hence just perform an Acquire load, but not have Release semantics. Using Acquire makes the store part of this operation Relaxed if it happens, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i64.

Migrating to compare_exchange and compare_exchange_weak

compare_and_swap is equivalent to compare_exchange with the following mapping for memory orderings:

Original	Success	Failure
Relaxed	Relaxed	Relaxed
Acquire	Acquire	Acquire
Release	Release	Relaxed
AcqRel	AcqRel	Acquire
SeqCst	SeqCst	SeqCst

compare_and_swap and compare_exchange also differ in their return type. You can use compare_exchange(...).unwrap_or_else(|x| x) to recover the behavior of compare_and_swap, but in most cases it is more idiomatic to check whether the return value is Ok or Err rather than to infer success vs failure based on the value that was read.

During migration, consider whether it makes sense to use compare_exchange_weak instead. compare_exchange_weak is allowed to fail spuriously even when the comparison succeeds, which allows the compiler to generate better assembly code when the compare and swap is used in a loop.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let some_var = AtomicI64::new(5);

assert_eq!(some_var.compare_and_swap(5, 10, Ordering::Relaxed), 5);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_and_swap(6, 12, Ordering::Relaxed), 10);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn compare_exchange( &self, current: i64, new: i64, success: Ordering, failure: Ordering, ) -> Result<i64, i64>

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is a result indicating whether the new value was written and containing the previous value. On success this value is guaranteed to be equal to current.

compare_exchange takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on i64.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let some_var = AtomicI64::new(5);

assert_eq!(some_var.compare_exchange(5, 10,
                                     Ordering::Acquire,
                                     Ordering::Relaxed),
           Ok(5));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_exchange(6, 12,
                                     Ordering::SeqCst,
                                     Ordering::Acquire),
           Err(10));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim! This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn compare_exchange_weak( &self, current: i64, new: i64, success: Ordering, failure: Ordering, ) -> Result<i64, i64>

Stores a value into the atomic integer if the current value is the same as the current value.

Unlike AtomicI64::compare_exchange, this function is allowed to spuriously fail even when the comparison succeeds, which can result in more efficient code on some platforms. The return value is a result indicating whether the new value was written and containing the previous value.

compare_exchange_weak takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on i64.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let val = AtomicI64::new(4);

let mut old = val.load(Ordering::Relaxed);
loop {
    let new = old * 2;
    match val.compare_exchange_weak(old, new, Ordering::SeqCst, Ordering::Relaxed) {
        Ok(_) => break,
        Err(x) => old = x,
    }
}

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim. This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn fetch_add(&self, val: i64, order: Ordering) -> i64

Adds to the current value, returning the previous value.

This operation wraps around on overflow.

fetch_add takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i64.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let foo = AtomicI64::new(0);
assert_eq!(foo.fetch_add(10, Ordering::SeqCst), 0);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_sub(&self, val: i64, order: Ordering) -> i64

Subtracts from the current value, returning the previous value.

This operation wraps around on overflow.

fetch_sub takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i64.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let foo = AtomicI64::new(20);
assert_eq!(foo.fetch_sub(10, Ordering::SeqCst), 20);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_and(&self, val: i64, order: Ordering) -> i64

Bitwise “and” with the current value.

Performs a bitwise “and” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_and takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i64.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let foo = AtomicI64::new(0b101101);
assert_eq!(foo.fetch_and(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b100001);

pub fn fetch_nand(&self, val: i64, order: Ordering) -> i64

Bitwise “nand” with the current value.

Performs a bitwise “nand” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_nand takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i64.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let foo = AtomicI64::new(0x13);
assert_eq!(foo.fetch_nand(0x31, Ordering::SeqCst), 0x13);
assert_eq!(foo.load(Ordering::SeqCst), !(0x13 & 0x31));

pub fn fetch_or(&self, val: i64, order: Ordering) -> i64

Bitwise “or” with the current value.

Performs a bitwise “or” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_or takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i64.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let foo = AtomicI64::new(0b101101);
assert_eq!(foo.fetch_or(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b111111);

pub fn fetch_xor(&self, val: i64, order: Ordering) -> i64

Bitwise “xor” with the current value.

Performs a bitwise “xor” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_xor takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i64.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let foo = AtomicI64::new(0b101101);
assert_eq!(foo.fetch_xor(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b011110);

pub fn fetch_update<F>( &self, set_order: Ordering, fetch_order: Ordering, f: F, ) -> Result<i64, i64>

where F: FnMut(i64) -> Option<i64>,

👎Deprecating in 1.99.0: renamed to try_update for consistency

An alias for AtomicI64::try_update .

pub fn try_update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(i64) -> Option<i64>, ) -> Result<i64, i64>

Fetches the value, and applies a function to it that returns an optional new value. Returns a Result of Ok(previous_value) if the function returned Some(_), else Err(previous_value).

See also: update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, as long as the function returns Some(_), but the function will have been applied only once to the stored value.

try_update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicI64::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on i64.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let x = AtomicI64::new(7);
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |_| None), Err(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(8));
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(i64) -> i64, ) -> i64

Fetches the value, applies a function to it that it return a new value. The new value is stored and the old value is returned.

See also: try_update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, but the function will have been applied only once to the stored value.

update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicI64::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on i64.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let x = AtomicI64::new(7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 8);
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn fetch_max(&self, val: i64, order: Ordering) -> i64

Maximum with the current value.

Finds the maximum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_max takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i64.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let foo = AtomicI64::new(23);
assert_eq!(foo.fetch_max(42, Ordering::SeqCst), 23);
assert_eq!(foo.load(Ordering::SeqCst), 42);

If you want to obtain the maximum value in one step, you can use the following:

use std::sync::atomic::{AtomicI64, Ordering};

let foo = AtomicI64::new(23);
let bar = 42;
let max_foo = foo.fetch_max(bar, Ordering::SeqCst).max(bar);
assert!(max_foo == 42);

pub fn fetch_min(&self, val: i64, order: Ordering) -> i64

Minimum with the current value.

Finds the minimum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_min takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on i64.

Examples

use std::sync::atomic::{AtomicI64, Ordering};

let foo = AtomicI64::new(23);
assert_eq!(foo.fetch_min(42, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 23);
assert_eq!(foo.fetch_min(22, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 22);

If you want to obtain the minimum value in one step, you can use the following:

use std::sync::atomic::{AtomicI64, Ordering};

let foo = AtomicI64::new(23);
let bar = 12;
let min_foo = foo.fetch_min(bar, Ordering::SeqCst).min(bar);
assert_eq!(min_foo, 12);

pub const fn as_ptr(&self) -> *mut i64

Returns a mutable pointer to the underlying integer.

Doing non-atomic reads and writes on the resulting integer can be a data race. This method is mostly useful for FFI, where the function signature may use *mut i64 instead of &AtomicI64.

Returning an *mut pointer from a shared reference to this atomic is safe because the atomic types work with interior mutability. All modifications of an atomic change the value through a shared reference, and can do so safely as long as they use atomic operations. Any use of the returned raw pointer requires an unsafe block and still has to uphold the requirements of the memory model.

Examples

use std::sync::atomic::AtomicI64;

extern "C" {
    fn my_atomic_op(arg: *mut i64);
}

let atomic = AtomicI64::new(1);

// SAFETY: Safe as long as `my_atomic_op` is atomic.
unsafe {
    my_atomic_op(atomic.as_ptr());
}

impl Atomic<u64>

pub const fn new(v: u64) -> Self

Creates a new atomic integer.

Examples

use std::sync::atomic::AtomicU64;

let atomic_forty_two = AtomicU64::new(42);

pub const unsafe fn from_ptr<'a>(ptr: *mut u64) -> &'a AtomicU64

Creates a new reference to an atomic integer from a pointer.

Examples

use std::sync::atomic::{self, AtomicU64};

// Get a pointer to an allocated value
let ptr: *mut u64 = Box::into_raw(Box::new(0));

assert!(ptr.cast::<AtomicU64>().is_aligned());

{
    // Create an atomic view of the allocated value
    let atomic = unsafe {AtomicU64::from_ptr(ptr) };

    // Use `atomic` for atomic operations, possibly share it with other threads
    atomic.store(1, atomic::Ordering::Relaxed);
}

// It's ok to non-atomically access the value behind `ptr`,
// since the reference to the atomic ended its lifetime in the block above
assert_eq!(unsafe { *ptr }, 1);

// Deallocate the value
unsafe { drop(Box::from_raw(ptr)) }

Safety

• ptr must be aligned to align_of::<AtomicU64>() (note that on some platforms this can be bigger than align_of::<u64>()).
• ptr must be valid for both reads and writes for the whole lifetime 'a.
• You must adhere to the Memory model for atomic accesses. In particular, it is not allowed to mix conflicting atomic and non-atomic accesses, or atomic accesses of different sizes, without synchronization.

pub fn get_mut(&mut self) -> &mut u64

Returns a mutable reference to the underlying integer.

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let mut some_var = AtomicU64::new(10);
assert_eq!(*some_var.get_mut(), 10);
*some_var.get_mut() = 5;
assert_eq!(some_var.load(Ordering::SeqCst), 5);

pub fn from_mut(v: &mut u64) -> &mut Self

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut u64.

Note: This function is only available on targets where AtomicU64 has the same alignment as u64.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicU64, Ordering};

let mut some_int = 123;
let a = AtomicU64::from_mut(&mut some_int);
a.store(100, Ordering::Relaxed);
assert_eq!(some_int, 100);

pub fn get_mut_slice(this: &mut [Self]) -> &mut [u64]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get non-atomic access to a &mut [AtomicU64] slice

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicU64, Ordering};

let mut some_ints = [const { AtomicU64::new(0) }; 10];

let view: &mut [u64] = AtomicU64::get_mut_slice(&mut some_ints);
assert_eq!(view, [0; 10]);
view
    .iter_mut()
    .enumerate()
    .for_each(|(idx, int)| *int = idx as _);

std::thread::scope(|s| {
    some_ints
        .iter()
        .enumerate()
        .for_each(|(idx, int)| {
            s.spawn(move || assert_eq!(int.load(Ordering::Relaxed), idx as _));
        })
});

pub fn from_mut_slice(v: &mut [u64]) -> &mut [Self]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut [u64] slice.

Note: This function is only available on targets where AtomicU64 has the same alignment as u64.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicU64, Ordering};

let mut some_ints = [0; 10];
let a = &*AtomicU64::from_mut_slice(&mut some_ints);
std::thread::scope(|s| {
    for i in 0..a.len() {
        s.spawn(move || a[i].store(i as _, Ordering::Relaxed));
    }
});
for (i, n) in some_ints.into_iter().enumerate() {
    assert_eq!(i, n as usize);
}

pub const fn into_inner(self) -> u64

Consumes the atomic and returns the contained value.

This is safe because passing self by value guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::AtomicU64;

let some_var = AtomicU64::new(5);
assert_eq!(some_var.into_inner(), 5);

pub fn load(&self, order: Ordering) -> u64

Loads a value from the atomic integer.

load takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Acquire and Relaxed.

Panics

Panics if order is Release or AcqRel.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let some_var = AtomicU64::new(5);

assert_eq!(some_var.load(Ordering::Relaxed), 5);

pub fn store(&self, val: u64, order: Ordering)

Stores a value into the atomic integer.

store takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Release and Relaxed.

Panics

Panics if order is Acquire or AcqRel.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let some_var = AtomicU64::new(5);

some_var.store(10, Ordering::Relaxed);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn swap(&self, val: u64, order: Ordering) -> u64

Stores a value into the atomic integer, returning the previous value.

swap takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u64.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let some_var = AtomicU64::new(5);

assert_eq!(some_var.swap(10, Ordering::Relaxed), 5);

pub fn compare_and_swap(&self, current: u64, new: u64, order: Ordering) -> u64

👎Deprecated since 1.50.0: Use compare_exchange or compare_exchange_weak instead

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is always the previous value. If it is equal to current, then the value was updated.

compare_and_swap also takes an Ordering argument which describes the memory ordering of this operation. Notice that even when using AcqRel, the operation might fail and hence just perform an Acquire load, but not have Release semantics. Using Acquire makes the store part of this operation Relaxed if it happens, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u64.

Migrating to compare_exchange and compare_exchange_weak

compare_and_swap is equivalent to compare_exchange with the following mapping for memory orderings:

Original	Success	Failure
Relaxed	Relaxed	Relaxed
Acquire	Acquire	Acquire
Release	Release	Relaxed
AcqRel	AcqRel	Acquire
SeqCst	SeqCst	SeqCst

compare_and_swap and compare_exchange also differ in their return type. You can use compare_exchange(...).unwrap_or_else(|x| x) to recover the behavior of compare_and_swap, but in most cases it is more idiomatic to check whether the return value is Ok or Err rather than to infer success vs failure based on the value that was read.

During migration, consider whether it makes sense to use compare_exchange_weak instead. compare_exchange_weak is allowed to fail spuriously even when the comparison succeeds, which allows the compiler to generate better assembly code when the compare and swap is used in a loop.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let some_var = AtomicU64::new(5);

assert_eq!(some_var.compare_and_swap(5, 10, Ordering::Relaxed), 5);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_and_swap(6, 12, Ordering::Relaxed), 10);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn compare_exchange( &self, current: u64, new: u64, success: Ordering, failure: Ordering, ) -> Result<u64, u64>

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is a result indicating whether the new value was written and containing the previous value. On success this value is guaranteed to be equal to current.

compare_exchange takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u64.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let some_var = AtomicU64::new(5);

assert_eq!(some_var.compare_exchange(5, 10,
                                     Ordering::Acquire,
                                     Ordering::Relaxed),
           Ok(5));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_exchange(6, 12,
                                     Ordering::SeqCst,
                                     Ordering::Acquire),
           Err(10));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim! This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn compare_exchange_weak( &self, current: u64, new: u64, success: Ordering, failure: Ordering, ) -> Result<u64, u64>

Stores a value into the atomic integer if the current value is the same as the current value.

Unlike AtomicU64::compare_exchange, this function is allowed to spuriously fail even when the comparison succeeds, which can result in more efficient code on some platforms. The return value is a result indicating whether the new value was written and containing the previous value.

compare_exchange_weak takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u64.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let val = AtomicU64::new(4);

let mut old = val.load(Ordering::Relaxed);
loop {
    let new = old * 2;
    match val.compare_exchange_weak(old, new, Ordering::SeqCst, Ordering::Relaxed) {
        Ok(_) => break,
        Err(x) => old = x,
    }
}

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim. This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn fetch_add(&self, val: u64, order: Ordering) -> u64

Adds to the current value, returning the previous value.

This operation wraps around on overflow.

fetch_add takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u64.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let foo = AtomicU64::new(0);
assert_eq!(foo.fetch_add(10, Ordering::SeqCst), 0);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_sub(&self, val: u64, order: Ordering) -> u64

Subtracts from the current value, returning the previous value.

This operation wraps around on overflow.

fetch_sub takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u64.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let foo = AtomicU64::new(20);
assert_eq!(foo.fetch_sub(10, Ordering::SeqCst), 20);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_and(&self, val: u64, order: Ordering) -> u64

Bitwise “and” with the current value.

Performs a bitwise “and” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_and takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u64.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let foo = AtomicU64::new(0b101101);
assert_eq!(foo.fetch_and(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b100001);

pub fn fetch_nand(&self, val: u64, order: Ordering) -> u64

Bitwise “nand” with the current value.

Performs a bitwise “nand” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_nand takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u64.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let foo = AtomicU64::new(0x13);
assert_eq!(foo.fetch_nand(0x31, Ordering::SeqCst), 0x13);
assert_eq!(foo.load(Ordering::SeqCst), !(0x13 & 0x31));

pub fn fetch_or(&self, val: u64, order: Ordering) -> u64

Bitwise “or” with the current value.

Performs a bitwise “or” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_or takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u64.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let foo = AtomicU64::new(0b101101);
assert_eq!(foo.fetch_or(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b111111);

pub fn fetch_xor(&self, val: u64, order: Ordering) -> u64

Bitwise “xor” with the current value.

Performs a bitwise “xor” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_xor takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u64.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let foo = AtomicU64::new(0b101101);
assert_eq!(foo.fetch_xor(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b011110);

pub fn fetch_update<F>( &self, set_order: Ordering, fetch_order: Ordering, f: F, ) -> Result<u64, u64>

where F: FnMut(u64) -> Option<u64>,

👎Deprecating in 1.99.0: renamed to try_update for consistency

An alias for AtomicU64::try_update .

pub fn try_update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(u64) -> Option<u64>, ) -> Result<u64, u64>

Fetches the value, and applies a function to it that returns an optional new value. Returns a Result of Ok(previous_value) if the function returned Some(_), else Err(previous_value).

See also: update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, as long as the function returns Some(_), but the function will have been applied only once to the stored value.

try_update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicU64::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u64.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let x = AtomicU64::new(7);
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |_| None), Err(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(8));
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(u64) -> u64, ) -> u64

Fetches the value, applies a function to it that it return a new value. The new value is stored and the old value is returned.

See also: try_update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, but the function will have been applied only once to the stored value.

update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicU64::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on u64.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let x = AtomicU64::new(7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 8);
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn fetch_max(&self, val: u64, order: Ordering) -> u64

Maximum with the current value.

Finds the maximum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_max takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u64.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let foo = AtomicU64::new(23);
assert_eq!(foo.fetch_max(42, Ordering::SeqCst), 23);
assert_eq!(foo.load(Ordering::SeqCst), 42);

If you want to obtain the maximum value in one step, you can use the following:

use std::sync::atomic::{AtomicU64, Ordering};

let foo = AtomicU64::new(23);
let bar = 42;
let max_foo = foo.fetch_max(bar, Ordering::SeqCst).max(bar);
assert!(max_foo == 42);

pub fn fetch_min(&self, val: u64, order: Ordering) -> u64

Minimum with the current value.

Finds the minimum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_min takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on u64.

Examples

use std::sync::atomic::{AtomicU64, Ordering};

let foo = AtomicU64::new(23);
assert_eq!(foo.fetch_min(42, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 23);
assert_eq!(foo.fetch_min(22, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 22);

If you want to obtain the minimum value in one step, you can use the following:

use std::sync::atomic::{AtomicU64, Ordering};

let foo = AtomicU64::new(23);
let bar = 12;
let min_foo = foo.fetch_min(bar, Ordering::SeqCst).min(bar);
assert_eq!(min_foo, 12);

pub const fn as_ptr(&self) -> *mut u64

Returns a mutable pointer to the underlying integer.

Doing non-atomic reads and writes on the resulting integer can be a data race. This method is mostly useful for FFI, where the function signature may use *mut u64 instead of &AtomicU64.

Returning an *mut pointer from a shared reference to this atomic is safe because the atomic types work with interior mutability. All modifications of an atomic change the value through a shared reference, and can do so safely as long as they use atomic operations. Any use of the returned raw pointer requires an unsafe block and still has to uphold the requirements of the memory model.

Examples

use std::sync::atomic::AtomicU64;

extern "C" {
    fn my_atomic_op(arg: *mut u64);
}

let atomic = AtomicU64::new(1);

// SAFETY: Safe as long as `my_atomic_op` is atomic.
unsafe {
    my_atomic_op(atomic.as_ptr());
}

impl Atomic<isize>

pub const fn new(v: isize) -> Self

Creates a new atomic integer.

Examples

use std::sync::atomic::AtomicIsize;

let atomic_forty_two = AtomicIsize::new(42);

pub const unsafe fn from_ptr<'a>(ptr: *mut isize) -> &'a AtomicIsize

Creates a new reference to an atomic integer from a pointer.

Examples

use std::sync::atomic::{self, AtomicIsize};

// Get a pointer to an allocated value
let ptr: *mut isize = Box::into_raw(Box::new(0));

assert!(ptr.cast::<AtomicIsize>().is_aligned());

{
    // Create an atomic view of the allocated value
    let atomic = unsafe {AtomicIsize::from_ptr(ptr) };

    // Use `atomic` for atomic operations, possibly share it with other threads
    atomic.store(1, atomic::Ordering::Relaxed);
}

// It's ok to non-atomically access the value behind `ptr`,
// since the reference to the atomic ended its lifetime in the block above
assert_eq!(unsafe { *ptr }, 1);

// Deallocate the value
unsafe { drop(Box::from_raw(ptr)) }

Safety

• ptr must be aligned to align_of::<AtomicIsize>() (note that on some platforms this can be bigger than align_of::<isize>()).
• ptr must be valid for both reads and writes for the whole lifetime 'a.
• You must adhere to the Memory model for atomic accesses. In particular, it is not allowed to mix conflicting atomic and non-atomic accesses, or atomic accesses of different sizes, without synchronization.

pub fn get_mut(&mut self) -> &mut isize

Returns a mutable reference to the underlying integer.

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let mut some_var = AtomicIsize::new(10);
assert_eq!(*some_var.get_mut(), 10);
*some_var.get_mut() = 5;
assert_eq!(some_var.load(Ordering::SeqCst), 5);

pub fn from_mut(v: &mut isize) -> &mut Self

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut isize.

Note: This function is only available on targets where AtomicIsize has the same alignment as isize.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicIsize, Ordering};

let mut some_int = 123;
let a = AtomicIsize::from_mut(&mut some_int);
a.store(100, Ordering::Relaxed);
assert_eq!(some_int, 100);

pub fn get_mut_slice(this: &mut [Self]) -> &mut [isize]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get non-atomic access to a &mut [AtomicIsize] slice

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicIsize, Ordering};

let mut some_ints = [const { AtomicIsize::new(0) }; 10];

let view: &mut [isize] = AtomicIsize::get_mut_slice(&mut some_ints);
assert_eq!(view, [0; 10]);
view
    .iter_mut()
    .enumerate()
    .for_each(|(idx, int)| *int = idx as _);

std::thread::scope(|s| {
    some_ints
        .iter()
        .enumerate()
        .for_each(|(idx, int)| {
            s.spawn(move || assert_eq!(int.load(Ordering::Relaxed), idx as _));
        })
});

pub fn from_mut_slice(v: &mut [isize]) -> &mut [Self]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut [isize] slice.

Note: This function is only available on targets where AtomicIsize has the same alignment as isize.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicIsize, Ordering};

let mut some_ints = [0; 10];
let a = &*AtomicIsize::from_mut_slice(&mut some_ints);
std::thread::scope(|s| {
    for i in 0..a.len() {
        s.spawn(move || a[i].store(i as _, Ordering::Relaxed));
    }
});
for (i, n) in some_ints.into_iter().enumerate() {
    assert_eq!(i, n as usize);
}

pub const fn into_inner(self) -> isize

Consumes the atomic and returns the contained value.

This is safe because passing self by value guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::AtomicIsize;

let some_var = AtomicIsize::new(5);
assert_eq!(some_var.into_inner(), 5);

pub fn load(&self, order: Ordering) -> isize

Loads a value from the atomic integer.

load takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Acquire and Relaxed.

Panics

Panics if order is Release or AcqRel.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let some_var = AtomicIsize::new(5);

assert_eq!(some_var.load(Ordering::Relaxed), 5);

pub fn store(&self, val: isize, order: Ordering)

Stores a value into the atomic integer.

store takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Release and Relaxed.

Panics

Panics if order is Acquire or AcqRel.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let some_var = AtomicIsize::new(5);

some_var.store(10, Ordering::Relaxed);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn swap(&self, val: isize, order: Ordering) -> isize

Stores a value into the atomic integer, returning the previous value.

swap takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on isize.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let some_var = AtomicIsize::new(5);

assert_eq!(some_var.swap(10, Ordering::Relaxed), 5);

pub fn compare_and_swap( &self, current: isize, new: isize, order: Ordering, ) -> isize

👎Deprecated since 1.50.0: Use compare_exchange or compare_exchange_weak instead

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is always the previous value. If it is equal to current, then the value was updated.

compare_and_swap also takes an Ordering argument which describes the memory ordering of this operation. Notice that even when using AcqRel, the operation might fail and hence just perform an Acquire load, but not have Release semantics. Using Acquire makes the store part of this operation Relaxed if it happens, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on isize.

Migrating to compare_exchange and compare_exchange_weak

compare_and_swap is equivalent to compare_exchange with the following mapping for memory orderings:

Original	Success	Failure
Relaxed	Relaxed	Relaxed
Acquire	Acquire	Acquire
Release	Release	Relaxed
AcqRel	AcqRel	Acquire
SeqCst	SeqCst	SeqCst

compare_and_swap and compare_exchange also differ in their return type. You can use compare_exchange(...).unwrap_or_else(|x| x) to recover the behavior of compare_and_swap, but in most cases it is more idiomatic to check whether the return value is Ok or Err rather than to infer success vs failure based on the value that was read.

During migration, consider whether it makes sense to use compare_exchange_weak instead. compare_exchange_weak is allowed to fail spuriously even when the comparison succeeds, which allows the compiler to generate better assembly code when the compare and swap is used in a loop.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let some_var = AtomicIsize::new(5);

assert_eq!(some_var.compare_and_swap(5, 10, Ordering::Relaxed), 5);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_and_swap(6, 12, Ordering::Relaxed), 10);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn compare_exchange( &self, current: isize, new: isize, success: Ordering, failure: Ordering, ) -> Result<isize, isize>

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is a result indicating whether the new value was written and containing the previous value. On success this value is guaranteed to be equal to current.

compare_exchange takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on isize.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let some_var = AtomicIsize::new(5);

assert_eq!(some_var.compare_exchange(5, 10,
                                     Ordering::Acquire,
                                     Ordering::Relaxed),
           Ok(5));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_exchange(6, 12,
                                     Ordering::SeqCst,
                                     Ordering::Acquire),
           Err(10));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim! This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn compare_exchange_weak( &self, current: isize, new: isize, success: Ordering, failure: Ordering, ) -> Result<isize, isize>

Stores a value into the atomic integer if the current value is the same as the current value.

Unlike AtomicIsize::compare_exchange, this function is allowed to spuriously fail even when the comparison succeeds, which can result in more efficient code on some platforms. The return value is a result indicating whether the new value was written and containing the previous value.

compare_exchange_weak takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on isize.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let val = AtomicIsize::new(4);

let mut old = val.load(Ordering::Relaxed);
loop {
    let new = old * 2;
    match val.compare_exchange_weak(old, new, Ordering::SeqCst, Ordering::Relaxed) {
        Ok(_) => break,
        Err(x) => old = x,
    }
}

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim. This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn fetch_add(&self, val: isize, order: Ordering) -> isize

Adds to the current value, returning the previous value.

This operation wraps around on overflow.

fetch_add takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on isize.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let foo = AtomicIsize::new(0);
assert_eq!(foo.fetch_add(10, Ordering::SeqCst), 0);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_sub(&self, val: isize, order: Ordering) -> isize

Subtracts from the current value, returning the previous value.

This operation wraps around on overflow.

fetch_sub takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on isize.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let foo = AtomicIsize::new(20);
assert_eq!(foo.fetch_sub(10, Ordering::SeqCst), 20);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_and(&self, val: isize, order: Ordering) -> isize

Bitwise “and” with the current value.

Performs a bitwise “and” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_and takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on isize.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let foo = AtomicIsize::new(0b101101);
assert_eq!(foo.fetch_and(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b100001);

pub fn fetch_nand(&self, val: isize, order: Ordering) -> isize

Bitwise “nand” with the current value.

Performs a bitwise “nand” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_nand takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on isize.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let foo = AtomicIsize::new(0x13);
assert_eq!(foo.fetch_nand(0x31, Ordering::SeqCst), 0x13);
assert_eq!(foo.load(Ordering::SeqCst), !(0x13 & 0x31));

pub fn fetch_or(&self, val: isize, order: Ordering) -> isize

Bitwise “or” with the current value.

Performs a bitwise “or” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_or takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on isize.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let foo = AtomicIsize::new(0b101101);
assert_eq!(foo.fetch_or(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b111111);

pub fn fetch_xor(&self, val: isize, order: Ordering) -> isize

Bitwise “xor” with the current value.

Performs a bitwise “xor” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_xor takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on isize.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let foo = AtomicIsize::new(0b101101);
assert_eq!(foo.fetch_xor(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b011110);

pub fn fetch_update<F>( &self, set_order: Ordering, fetch_order: Ordering, f: F, ) -> Result<isize, isize>

where F: FnMut(isize) -> Option<isize>,

👎Deprecating in 1.99.0: renamed to try_update for consistency

An alias for AtomicIsize::try_update .

pub fn try_update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(isize) -> Option<isize>, ) -> Result<isize, isize>

Fetches the value, and applies a function to it that returns an optional new value. Returns a Result of Ok(previous_value) if the function returned Some(_), else Err(previous_value).

See also: update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, as long as the function returns Some(_), but the function will have been applied only once to the stored value.

try_update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicIsize::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on isize.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let x = AtomicIsize::new(7);
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |_| None), Err(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(8));
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(isize) -> isize, ) -> isize

Fetches the value, applies a function to it that it return a new value. The new value is stored and the old value is returned.

See also: try_update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, but the function will have been applied only once to the stored value.

update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicIsize::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on isize.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let x = AtomicIsize::new(7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 8);
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn fetch_max(&self, val: isize, order: Ordering) -> isize

Maximum with the current value.

Finds the maximum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_max takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on isize.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let foo = AtomicIsize::new(23);
assert_eq!(foo.fetch_max(42, Ordering::SeqCst), 23);
assert_eq!(foo.load(Ordering::SeqCst), 42);

If you want to obtain the maximum value in one step, you can use the following:

use std::sync::atomic::{AtomicIsize, Ordering};

let foo = AtomicIsize::new(23);
let bar = 42;
let max_foo = foo.fetch_max(bar, Ordering::SeqCst).max(bar);
assert!(max_foo == 42);

pub fn fetch_min(&self, val: isize, order: Ordering) -> isize

Minimum with the current value.

Finds the minimum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_min takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on isize.

Examples

use std::sync::atomic::{AtomicIsize, Ordering};

let foo = AtomicIsize::new(23);
assert_eq!(foo.fetch_min(42, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 23);
assert_eq!(foo.fetch_min(22, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 22);

If you want to obtain the minimum value in one step, you can use the following:

use std::sync::atomic::{AtomicIsize, Ordering};

let foo = AtomicIsize::new(23);
let bar = 12;
let min_foo = foo.fetch_min(bar, Ordering::SeqCst).min(bar);
assert_eq!(min_foo, 12);

pub const fn as_ptr(&self) -> *mut isize

Returns a mutable pointer to the underlying integer.

Doing non-atomic reads and writes on the resulting integer can be a data race. This method is mostly useful for FFI, where the function signature may use *mut isize instead of &AtomicIsize.

Returning an *mut pointer from a shared reference to this atomic is safe because the atomic types work with interior mutability. All modifications of an atomic change the value through a shared reference, and can do so safely as long as they use atomic operations. Any use of the returned raw pointer requires an unsafe block and still has to uphold the requirements of the memory model.

Examples

use std::sync::atomic::AtomicIsize;

extern "C" {
    fn my_atomic_op(arg: *mut isize);
}

let atomic = AtomicIsize::new(1);

// SAFETY: Safe as long as `my_atomic_op` is atomic.
unsafe {
    my_atomic_op(atomic.as_ptr());
}

impl Atomic<usize>

pub const fn new(v: usize) -> Self

Creates a new atomic integer.

Examples

use std::sync::atomic::AtomicUsize;

let atomic_forty_two = AtomicUsize::new(42);

pub const unsafe fn from_ptr<'a>(ptr: *mut usize) -> &'a AtomicUsize

Creates a new reference to an atomic integer from a pointer.

Examples

use std::sync::atomic::{self, AtomicUsize};

// Get a pointer to an allocated value
let ptr: *mut usize = Box::into_raw(Box::new(0));

assert!(ptr.cast::<AtomicUsize>().is_aligned());

{
    // Create an atomic view of the allocated value
    let atomic = unsafe {AtomicUsize::from_ptr(ptr) };

    // Use `atomic` for atomic operations, possibly share it with other threads
    atomic.store(1, atomic::Ordering::Relaxed);
}

// It's ok to non-atomically access the value behind `ptr`,
// since the reference to the atomic ended its lifetime in the block above
assert_eq!(unsafe { *ptr }, 1);

// Deallocate the value
unsafe { drop(Box::from_raw(ptr)) }

Safety

• ptr must be aligned to align_of::<AtomicUsize>() (note that on some platforms this can be bigger than align_of::<usize>()).
• ptr must be valid for both reads and writes for the whole lifetime 'a.
• You must adhere to the Memory model for atomic accesses. In particular, it is not allowed to mix conflicting atomic and non-atomic accesses, or atomic accesses of different sizes, without synchronization.

pub fn get_mut(&mut self) -> &mut usize

Returns a mutable reference to the underlying integer.

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let mut some_var = AtomicUsize::new(10);
assert_eq!(*some_var.get_mut(), 10);
*some_var.get_mut() = 5;
assert_eq!(some_var.load(Ordering::SeqCst), 5);

pub fn from_mut(v: &mut usize) -> &mut Self

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut usize.

Note: This function is only available on targets where AtomicUsize has the same alignment as usize.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicUsize, Ordering};

let mut some_int = 123;
let a = AtomicUsize::from_mut(&mut some_int);
a.store(100, Ordering::Relaxed);
assert_eq!(some_int, 100);

pub fn get_mut_slice(this: &mut [Self]) -> &mut [usize]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get non-atomic access to a &mut [AtomicUsize] slice

This is safe because the mutable reference guarantees that no other threads are concurrently accessing the atomic data.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicUsize, Ordering};

let mut some_ints = [const { AtomicUsize::new(0) }; 10];

let view: &mut [usize] = AtomicUsize::get_mut_slice(&mut some_ints);
assert_eq!(view, [0; 10]);
view
    .iter_mut()
    .enumerate()
    .for_each(|(idx, int)| *int = idx as _);

std::thread::scope(|s| {
    some_ints
        .iter()
        .enumerate()
        .for_each(|(idx, int)| {
            s.spawn(move || assert_eq!(int.load(Ordering::Relaxed), idx as _));
        })
});

pub fn from_mut_slice(v: &mut [usize]) -> &mut [Self]

🔬This is a nightly-only experimental API. (atomic_from_mut #76314)

Get atomic access to a &mut [usize] slice.

Note: This function is only available on targets where AtomicUsize has the same alignment as usize.

Examples

#![feature(atomic_from_mut)]
use std::sync::atomic::{AtomicUsize, Ordering};

let mut some_ints = [0; 10];
let a = &*AtomicUsize::from_mut_slice(&mut some_ints);
std::thread::scope(|s| {
    for i in 0..a.len() {
        s.spawn(move || a[i].store(i as _, Ordering::Relaxed));
    }
});
for (i, n) in some_ints.into_iter().enumerate() {
    assert_eq!(i, n as usize);
}

pub const fn into_inner(self) -> usize

Consumes the atomic and returns the contained value.

This is safe because passing self by value guarantees that no other threads are concurrently accessing the atomic data.

Examples

use std::sync::atomic::AtomicUsize;

let some_var = AtomicUsize::new(5);
assert_eq!(some_var.into_inner(), 5);

pub fn load(&self, order: Ordering) -> usize

Loads a value from the atomic integer.

load takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Acquire and Relaxed.

Panics

Panics if order is Release or AcqRel.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let some_var = AtomicUsize::new(5);

assert_eq!(some_var.load(Ordering::Relaxed), 5);

pub fn store(&self, val: usize, order: Ordering)

Stores a value into the atomic integer.

store takes an Ordering argument which describes the memory ordering of this operation. Possible values are SeqCst, Release and Relaxed.

Panics

Panics if order is Acquire or AcqRel.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let some_var = AtomicUsize::new(5);

some_var.store(10, Ordering::Relaxed);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn swap(&self, val: usize, order: Ordering) -> usize

Stores a value into the atomic integer, returning the previous value.

swap takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on usize.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let some_var = AtomicUsize::new(5);

assert_eq!(some_var.swap(10, Ordering::Relaxed), 5);

pub fn compare_and_swap( &self, current: usize, new: usize, order: Ordering, ) -> usize

👎Deprecated since 1.50.0: Use compare_exchange or compare_exchange_weak instead

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is always the previous value. If it is equal to current, then the value was updated.

compare_and_swap also takes an Ordering argument which describes the memory ordering of this operation. Notice that even when using AcqRel, the operation might fail and hence just perform an Acquire load, but not have Release semantics. Using Acquire makes the store part of this operation Relaxed if it happens, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on usize.

Migrating to compare_exchange and compare_exchange_weak

compare_and_swap is equivalent to compare_exchange with the following mapping for memory orderings:

Original	Success	Failure
Relaxed	Relaxed	Relaxed
Acquire	Acquire	Acquire
Release	Release	Relaxed
AcqRel	AcqRel	Acquire
SeqCst	SeqCst	SeqCst

compare_and_swap and compare_exchange also differ in their return type. You can use compare_exchange(...).unwrap_or_else(|x| x) to recover the behavior of compare_and_swap, but in most cases it is more idiomatic to check whether the return value is Ok or Err rather than to infer success vs failure based on the value that was read.

During migration, consider whether it makes sense to use compare_exchange_weak instead. compare_exchange_weak is allowed to fail spuriously even when the comparison succeeds, which allows the compiler to generate better assembly code when the compare and swap is used in a loop.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let some_var = AtomicUsize::new(5);

assert_eq!(some_var.compare_and_swap(5, 10, Ordering::Relaxed), 5);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_and_swap(6, 12, Ordering::Relaxed), 10);
assert_eq!(some_var.load(Ordering::Relaxed), 10);

pub fn compare_exchange( &self, current: usize, new: usize, success: Ordering, failure: Ordering, ) -> Result<usize, usize>

Stores a value into the atomic integer if the current value is the same as the current value.

The return value is a result indicating whether the new value was written and containing the previous value. On success this value is guaranteed to be equal to current.

compare_exchange takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on usize.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let some_var = AtomicUsize::new(5);

assert_eq!(some_var.compare_exchange(5, 10,
                                     Ordering::Acquire,
                                     Ordering::Relaxed),
           Ok(5));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

assert_eq!(some_var.compare_exchange(6, 12,
                                     Ordering::SeqCst,
                                     Ordering::Acquire),
           Err(10));
assert_eq!(some_var.load(Ordering::Relaxed), 10);

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim! This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn compare_exchange_weak( &self, current: usize, new: usize, success: Ordering, failure: Ordering, ) -> Result<usize, usize>

Stores a value into the atomic integer if the current value is the same as the current value.

Unlike AtomicUsize::compare_exchange, this function is allowed to spuriously fail even when the comparison succeeds, which can result in more efficient code on some platforms. The return value is a result indicating whether the new value was written and containing the previous value.

compare_exchange_weak takes two Ordering arguments to describe the memory ordering of this operation. success describes the required ordering for the read-modify-write operation that takes place if the comparison with current succeeds. failure describes the required ordering for the load operation that takes place when the comparison fails. Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the successful load Relaxed. The failure ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on usize.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let val = AtomicUsize::new(4);

let mut old = val.load(Ordering::Relaxed);
loop {
    let new = old * 2;
    match val.compare_exchange_weak(old, new, Ordering::SeqCst, Ordering::Relaxed) {
        Ok(_) => break,
        Err(x) => old = x,
    }
}

Considerations

compare_exchange is a compare-and-swap operation and thus exhibits the usual downsides of CAS operations. In particular, a load of the value followed by a successful compare_exchange with the previous load does not ensure that other threads have not changed the value in the interim. This is usually important when the equality check in the compare_exchange is being used to check the identity of a value, but equality does not necessarily imply identity. This is a particularly common case for pointers, as a pointer holding the same address does not imply that the same object exists at that address! In this case, compare_exchange can lead to the ABA problem.

pub fn fetch_add(&self, val: usize, order: Ordering) -> usize

Adds to the current value, returning the previous value.

This operation wraps around on overflow.

fetch_add takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on usize.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let foo = AtomicUsize::new(0);
assert_eq!(foo.fetch_add(10, Ordering::SeqCst), 0);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_sub(&self, val: usize, order: Ordering) -> usize

Subtracts from the current value, returning the previous value.

This operation wraps around on overflow.

fetch_sub takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on usize.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let foo = AtomicUsize::new(20);
assert_eq!(foo.fetch_sub(10, Ordering::SeqCst), 20);
assert_eq!(foo.load(Ordering::SeqCst), 10);

pub fn fetch_and(&self, val: usize, order: Ordering) -> usize

Bitwise “and” with the current value.

Performs a bitwise “and” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_and takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on usize.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let foo = AtomicUsize::new(0b101101);
assert_eq!(foo.fetch_and(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b100001);

pub fn fetch_nand(&self, val: usize, order: Ordering) -> usize

Bitwise “nand” with the current value.

Performs a bitwise “nand” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_nand takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on usize.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let foo = AtomicUsize::new(0x13);
assert_eq!(foo.fetch_nand(0x31, Ordering::SeqCst), 0x13);
assert_eq!(foo.load(Ordering::SeqCst), !(0x13 & 0x31));

pub fn fetch_or(&self, val: usize, order: Ordering) -> usize

Bitwise “or” with the current value.

Performs a bitwise “or” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_or takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on usize.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let foo = AtomicUsize::new(0b101101);
assert_eq!(foo.fetch_or(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b111111);

pub fn fetch_xor(&self, val: usize, order: Ordering) -> usize

Bitwise “xor” with the current value.

Performs a bitwise “xor” operation on the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_xor takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on usize.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let foo = AtomicUsize::new(0b101101);
assert_eq!(foo.fetch_xor(0b110011, Ordering::SeqCst), 0b101101);
assert_eq!(foo.load(Ordering::SeqCst), 0b011110);

pub fn fetch_update<F>( &self, set_order: Ordering, fetch_order: Ordering, f: F, ) -> Result<usize, usize>

where F: FnMut(usize) -> Option<usize>,

👎Deprecating in 1.99.0: renamed to try_update for consistency

An alias for AtomicUsize::try_update .

pub fn try_update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(usize) -> Option<usize>, ) -> Result<usize, usize>

Fetches the value, and applies a function to it that returns an optional new value. Returns a Result of Ok(previous_value) if the function returned Some(_), else Err(previous_value).

See also: update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, as long as the function returns Some(_), but the function will have been applied only once to the stored value.

try_update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicUsize::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on usize.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let x = AtomicUsize::new(7);
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |_| None), Err(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(7));
assert_eq!(x.try_update(Ordering::SeqCst, Ordering::SeqCst, |x| Some(x + 1)), Ok(8));
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn update( &self, set_order: Ordering, fetch_order: Ordering, f: impl FnMut(usize) -> usize, ) -> usize

Fetches the value, applies a function to it that it return a new value. The new value is stored and the old value is returned.

See also: try_update.

Note: This may call the function multiple times if the value has been changed from other threads in the meantime, but the function will have been applied only once to the stored value.

update takes two Ordering arguments to describe the memory ordering of this operation. The first describes the required ordering for when the operation finally succeeds while the second describes the required ordering for loads. These correspond to the success and failure orderings of AtomicUsize::compare_exchange respectively.

Using Acquire as success ordering makes the store part of this operation Relaxed, and using Release makes the final successful load Relaxed. The (failed) load ordering can only be SeqCst, Acquire or Relaxed.

Note: This method is only available on platforms that support atomic operations on usize.

Considerations

This method is not magic; it is not provided by the hardware, and does not act like a critical section or mutex.

It is implemented on top of an atomic compare-and-swap operation, and thus is subject to the usual drawbacks of CAS operations. In particular, be careful of the ABA problem if this atomic integer is an index or more generally if knowledge of only the bitwise value of the atomic is not in and of itself sufficient to ensure any required preconditions.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let x = AtomicUsize::new(7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 7);
assert_eq!(x.update(Ordering::SeqCst, Ordering::SeqCst, |x| x + 1), 8);
assert_eq!(x.load(Ordering::SeqCst), 9);

pub fn fetch_max(&self, val: usize, order: Ordering) -> usize

Maximum with the current value.

Finds the maximum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_max takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on usize.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let foo = AtomicUsize::new(23);
assert_eq!(foo.fetch_max(42, Ordering::SeqCst), 23);
assert_eq!(foo.load(Ordering::SeqCst), 42);

If you want to obtain the maximum value in one step, you can use the following:

use std::sync::atomic::{AtomicUsize, Ordering};

let foo = AtomicUsize::new(23);
let bar = 42;
let max_foo = foo.fetch_max(bar, Ordering::SeqCst).max(bar);
assert!(max_foo == 42);

pub fn fetch_min(&self, val: usize, order: Ordering) -> usize

Minimum with the current value.

Finds the minimum of the current value and the argument val, and sets the new value to the result.

Returns the previous value.

fetch_min takes an Ordering argument which describes the memory ordering of this operation. All ordering modes are possible. Note that using Acquire makes the store part of this operation Relaxed, and using Release makes the load part Relaxed.

Note: This method is only available on platforms that support atomic operations on usize.

Examples

use std::sync::atomic::{AtomicUsize, Ordering};

let foo = AtomicUsize::new(23);
assert_eq!(foo.fetch_min(42, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 23);
assert_eq!(foo.fetch_min(22, Ordering::Relaxed), 23);
assert_eq!(foo.load(Ordering::Relaxed), 22);

If you want to obtain the minimum value in one step, you can use the following:

use std::sync::atomic::{AtomicUsize, Ordering};

let foo = AtomicUsize::new(23);
let bar = 12;
let min_foo = foo.fetch_min(bar, Ordering::SeqCst).min(bar);
assert_eq!(min_foo, 12);

pub const fn as_ptr(&self) -> *mut usize

Returns a mutable pointer to the underlying integer.

Doing non-atomic reads and writes on the resulting integer can be a data race. This method is mostly useful for FFI, where the function signature may use *mut usize instead of &AtomicUsize.

Returning an *mut pointer from a shared reference to this atomic is safe because the atomic types work with interior mutability. All modifications of an atomic change the value through a shared reference, and can do so safely as long as they use atomic operations. Any use of the returned raw pointer requires an unsafe block and still has to uphold the requirements of the memory model.

Examples

use std::sync::atomic::AtomicUsize;

extern "C" {
    fn my_atomic_op(arg: *mut usize);
}

let atomic = AtomicUsize::new(1);

// SAFETY: Safe as long as `my_atomic_op` is atomic.
unsafe {
    my_atomic_op(atomic.as_ptr());
}

Trait Implementations

impl<T: AtomicPrimitive> Send for Atomic<T>

impl<T: AtomicPrimitive> Sync for Atomic<T>