Unverified Commit a09ee81e authored by Stefan Lankes's avatar Stefan Lankes Committed by GitHub
Browse files

add generic to align types (#71)

it is derived from crossbeam-utils, the main difference is that
this version provides a constant initialization routine
parent 91449b7b
use core::fmt;
use core::ops::{Deref, DerefMut};
/// Pads and aligns a value to the length of a cache line.
///
/// This version is derived from crossbeam-utils and support "const initialization routine".
/// In concurrent programming, sometimes it is desirable to make sure commonly accessed pieces of
/// data are not placed into the same cache line. Updating an atomic value invalides the whole
/// cache line it belongs to, which makes the next access to the same cache line slower for other
/// CPU cores. Use `CachePadded` to ensure updating one piece of data doesn't invalidate other
/// cached data.
///
/// # Size and alignment
///
/// Cache lines are assumed to be N bytes long, depending on the architecture:
///
/// * On x86-64, N = 128.
/// * On all others, N = 64.
///
/// Note that N is just a reasonable guess and is not guaranteed to match the actual cache line
/// length of the machine the program is running on. On modern Intel architectures, spatial
/// prefetcher is pulling pairs of 64-byte cache lines at a time, so we pessimistically assume that
/// cache lines are 128 bytes long.
///
/// The size of `CachePadded<T>` is the smallest multiple of N bytes large enough to accommodate
/// a value of type `T`.
///
/// The alignment of `CachePadded<T>` is the maximum of N bytes and the alignment of `T`.
///
/// # Examples
///
/// Alignment and padding:
///
/// ```
/// use crossbeam_utils::CachePadded;
///
/// let array = [CachePadded::new(1i8), CachePadded::new(2i8)];
/// let addr1 = &*array[0] as *const i8 as usize;
/// let addr2 = &*array[1] as *const i8 as usize;
///
/// assert!(addr2 - addr1 >= 64);
/// assert_eq!(addr1 % 64, 0);
/// assert_eq!(addr2 % 64, 0);
/// ```
///
/// When building a concurrent queue with a head and a tail index, it is wise to place them in
/// different cache lines so that concurrent threads pushing and popping elements don't invalidate
/// each other's cache lines:
///
/// ```
/// use crossbeam_utils::CachePadded;
/// use std::sync::atomic::AtomicUsize;
///
/// struct Queue<T> {
/// head: CachePadded<AtomicUsize>,
/// tail: CachePadded<AtomicUsize>,
/// buffer: *mut T,
/// }
/// ```
#[derive(Clone, Copy, Default, Hash, PartialEq, Eq)]
// Starting from Intel's Sandy Bridge, spatial prefetcher is now pulling pairs of 64-byte cache
// lines at a time, so we have to align to 128 bytes rather than 64.
//
// Sources:
// - https://www.intel.com/content/dam/www/public/us/en/documents/manuals/64-ia-32-architectures-optimization-manual.pdf
// - https://github.com/facebook/folly/blob/1b5288e6eea6df074758f877c849b6e73bbb9fbb/folly/lang/Align.h#L107
#[cfg_attr(target_arch = "x86_64", repr(align(128)))]
#[cfg_attr(not(target_arch = "x86_64"), repr(align(64)))]
pub struct CachePadded<T> {
value: T,
}
unsafe impl<T: Send> Send for CachePadded<T> {}
unsafe impl<T: Sync> Sync for CachePadded<T> {}
impl<T> CachePadded<T> {
/// Pads and aligns a value to the length of a cache line.
///
/// # Examples
///
/// ```
/// use crossbeam_utils::CachePadded;
///
/// let padded_value = CachePadded::new(1);
/// ```
pub const fn new(t: T) -> CachePadded<T> {
CachePadded::<T> { value: t }
}
/// Returns the inner value.
///
/// # Examples
///
/// ```
/// use crossbeam_utils::CachePadded;
///
/// let padded_value = CachePadded::new(7);
/// let value = padded_value.into_inner();
/// assert_eq!(value, 7);
/// ```
pub fn into_inner(self) -> T {
self.value
}
}
impl<T> Deref for CachePadded<T> {
type Target = T;
fn deref(&self) -> &T {
&self.value
}
}
impl<T> DerefMut for CachePadded<T> {
fn deref_mut(&mut self) -> &mut T {
&mut self.value
}
}
impl<T: fmt::Debug> fmt::Debug for CachePadded<T> {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
f.debug_struct("CachePadded")
.field("value", &self.value)
.finish()
}
}
impl<T> From<T> for CachePadded<T> {
fn from(t: T) -> Self {
CachePadded::new(t)
}
}
......@@ -7,8 +7,10 @@
use crate::arch::irq;
mod cachepadded;
mod doublylinkedlist;
pub use self::cachepadded::*;
pub use self::doublylinkedlist::*;
pub struct AvoidInterrupts(bool);
......
......@@ -6,6 +6,7 @@
// copied, modified, or distributed except according to those terms.
use crate::arch::irq;
use crate::collections::CachePadded;
use core::cell::UnsafeCell;
use core::fmt;
use core::marker::Sync;
......@@ -45,8 +46,8 @@ use core::sync::atomic::{spin_loop_hint, AtomicBool, AtomicUsize, Ordering};
/// assert_eq!(answer, 2);
/// ```
pub struct Spinlock<T: ?Sized> {
queue: AtomicUsize,
dequeue: AtomicUsize,
queue: CachePadded<AtomicUsize>,
dequeue: CachePadded<AtomicUsize>,
data: UnsafeCell<T>,
}
......@@ -55,7 +56,7 @@ pub struct Spinlock<T: ?Sized> {
/// When the guard falls out of scope it will release the lock.
pub struct SpinlockGuard<'a, T: ?Sized + 'a> {
//queue: &'a AtomicUsize,
dequeue: &'a AtomicUsize,
dequeue: &'a CachePadded<AtomicUsize>,
data: &'a mut T,
}
......@@ -66,8 +67,8 @@ unsafe impl<T: ?Sized> Send for Spinlock<T> {}
impl<T> Spinlock<T> {
pub const fn new(user_data: T) -> Spinlock<T> {
Spinlock {
queue: AtomicUsize::new(0),
dequeue: AtomicUsize::new(1),
queue: CachePadded::new(AtomicUsize::new(0)),
dequeue: CachePadded::new(AtomicUsize::new(1)),
data: UnsafeCell::new(user_data),
}
}
......@@ -167,8 +168,8 @@ impl<'a, T: ?Sized> Drop for SpinlockGuard<'a, T> {
/// assert_eq!(answer, 2);
/// ```
pub struct SpinlockIrqSave<T: ?Sized> {
queue: AtomicUsize,
dequeue: AtomicUsize,
queue: CachePadded<AtomicUsize>,
dequeue: CachePadded<AtomicUsize>,
irq: AtomicBool,
data: UnsafeCell<T>,
}
......@@ -178,7 +179,7 @@ pub struct SpinlockIrqSave<T: ?Sized> {
/// When the guard falls out of scope it will release the lock.
pub struct SpinlockIrqSaveGuard<'a, T: ?Sized + 'a> {
//queue: &'a AtomicUsize,
dequeue: &'a AtomicUsize,
dequeue: &'a CachePadded<AtomicUsize>,
irq: &'a AtomicBool,
data: &'a mut T,
}
......@@ -190,8 +191,8 @@ unsafe impl<T: ?Sized> Send for SpinlockIrqSave<T> {}
impl<T> SpinlockIrqSave<T> {
pub const fn new(user_data: T) -> SpinlockIrqSave<T> {
SpinlockIrqSave {
queue: AtomicUsize::new(0),
dequeue: AtomicUsize::new(1),
queue: CachePadded::new(AtomicUsize::new(0)),
dequeue: CachePadded::new(AtomicUsize::new(1)),
irq: AtomicBool::new(false),
data: UnsafeCell::new(user_data),
}
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment