Struct CachePadded 

#[repr(align(64))]
pub struct CachePadded<T>(pub T);

A wrapper that aligns T to a cache-line boundary.

On the ET-SoC-1 (a software-coherent architecture), two values sharing a cache line that are written by distinct harts without explicit cache operations cause false-sharing corruption. Wrapping per-hart output data in CachePadded ensures each instance occupies a distinct 64-byte line, making cross-hart false sharing structurally impossible regardless of the surrounding allocation layout.

The inner value is accessed directly via the public tuple field 0.

Example

use et_abi::CachePadded;
let cell: CachePadded<u64> = CachePadded(0);
assert_eq!(core::mem::align_of::<CachePadded<u64>>(), 64);

Tuple Fields

0: T

Trait Implementations

impl<T: Clone> Clone for CachePadded<T>

fn clone(&self) -> CachePadded<T>

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl<T: Copy> Copy for CachePadded<T>

impl<T: Debug> Debug for CachePadded<T>

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<T: Default> Default for CachePadded<T>

fn default() -> CachePadded<T>

Returns the “default value” for a type.

impl<T: Eq> Eq for CachePadded<T>

impl<T: PartialEq> PartialEq for CachePadded<T>

fn eq(&self, other: &CachePadded<T>) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl<T: PartialEq> StructuralPartialEq for CachePadded<T>