Struct crossbeam_utils::cache_padded::CachePadded [] [src]

pub struct CachePadded<T> { /* fields omitted */ }

Pads T to the length of a cache line.

Sometimes concurrent programming requires a piece of data to be padded out to the size of a cacheline to avoid "false sharing": cache lines being invalidated due to unrelated concurrent activity. Use this type when you want to avoid cache locality.

At the moment, cache lines are assumed to be 64 bytes on all architectures.

Size and alignment

By default, the size of CachePadded<T> is 64 bytes. If T is larger than that, then CachePadded::<T>::new will panic. Alignment of CachePadded<T> is the same as that of T.

However, if the nightly feature is enabled, arbitrarily large types T can be stored inside a CachePadded<T>. The size will then be a multiple of 64 at least the size of T, and the alignment will be the maximum of 64 and the alignment of T.

Methods

impl<T> CachePadded<T>
[src]

[src]

Pads a value to the length of a cache line.

Panics

If nightly is not enabled and T is larger than 64 bytes, this function will panic.

Trait Implementations

impl<T> Drop for CachePadded<T>
[src]

[src]

Executes the destructor for this type. Read more

impl<T: Send> Send for CachePadded<T>
[src]

impl<T: Sync> Sync for CachePadded<T>
[src]

impl<T> Deref for CachePadded<T>
[src]

The resulting type after dereferencing.

[src]

Dereferences the value.

impl<T> DerefMut for CachePadded<T>
[src]

[src]

Mutably dereferences the value.

impl<T: Default> Default for CachePadded<T>
[src]

[src]

Returns the "default value" for a type. Read more

impl<T: Clone> Clone for CachePadded<T>
[src]

[src]

Returns a copy of the value. Read more

1.0.0
[src]

Performs copy-assignment from source. Read more

impl<T: Debug> Debug for CachePadded<T>
[src]

[src]

Formats the value using the given formatter. Read more

impl<T> From<T> for CachePadded<T>
[src]

[src]

Performs the conversion.