Skip to content

Latest commit

 

History

History
197 lines (149 loc) · 7.18 KB

readme.md

File metadata and controls

197 lines (149 loc) · 7.18 KB

Slice Deque

crates.io version Travis build status Appveyor build status Codecov code coverage Docs License

A double-ended queue that Derefs into a slice, also known as a ring buffer or circular buffer.

Advantages

The main advantages of SliceDeque are:

  • nicer API: since it Derefs to a slice, all operations that work on slices (like sort) "just work" for SliceDeque.

  • efficient: as efficient as a slice (iteration, sorting, etc.), more efficient in general than VecDeque.

Platform Support

Windows, Linux, MacOS and every other unix-like OS is supported (although maybe untested). The following targets are known to work and pass all tests:

Linux

  • aarch64-unknown-linux-gnu
  • arm-unknown-linux-gnueabi
  • arm-unknown-linux-musleabi
  • armv7-unknown-linux-gnueabihf
  • armv7-unknown-linux-musleabihf
  • i586-unknown-linux-gnu
  • i686-unknown-linux-gnu
  • i686-unknown-linux-musl
  • mips-unknown-linux-gnu
  • mips64-unknown-linux-gnuabi64
  • mips64el-unknown-linux-gnuabi64
  • mipsel-unknown-linux-gnu
  • powerpc-unknown-linux-gnu
  • powerpc64-unknown-linux-gnu
  • powerpc64le-unknown-linux-gnu
  • x86_64-unknown-linux-gnu
  • x86_64-unknown-linux-musl
  • aarch64-linux-android
  • arm-linux-androideabi
  • armv7-linux-androideabi
  • x86_64-linux-android

MacOS X

  • i686-apple-darwin
  • x86_64-apple-darwin

Windows

  • x86_64-pc-windows-msvc

Drawbacks

The main drawbacks of SliceDeque are:

  • "constrained" platform support: the operating system must support virtual memory. In general, if you can use std, you can use SliceDeque.

  • global allocator bypass: SliceDeque bypasses Rust's global allocator / it is its own memory allocator, talking directly to the OS. That is, allocating and growing SliceDeques always involve system calls, while a VecDeque backed-up by a global allocator might receive memory owned by the allocator without any system calls at all.

  • smallest capacity constrained by the allocation granularity of the OS: some operating systems allow SliceDeque to allocate memory in 4/8/64 kB chunks.

When shouldn't you use it? In my opinion, if

  • you need to target #[no_std], or
  • you can't use it (because your platform doesn't support it)

you must use something else. If.

  • your ring-buffer's are very small,

then by using SliceDeque you might be trading memory for performance. Also,

  • your application has many short-lived ring-buffers,

the cost of the system calls required to set up and grow the SliceDeques might not be amortized by your application (update: there is a pull-request open that caches allocations in thread-local heaps when the feature use_std is enabled significantly improving the performance of short-lived ring-buffers, but it has not been merged yet). Whether any of these trade-offs are worth it or not is application dependent, so don't take my word for it: measure.

How it works

The double-ended queue in the standard library (VecDeque) is implemented using a growable ring buffer (0 represents uninitialized memory, and T represents one element in the queue):

// [ 0 | 0 | 0 | T | T | T | 0 ]
//               ^:head  ^:tail

When the queue grows beyond the end of the allocated buffer, its tail wraps around:

// [ T | T | 0 | T | T | T | T ]
//       ^:tail  ^:head

As a consequence, VecDeque cannot Deref into a slice, since its elements do not, in general, occupy a contiguous memory region. This complicates the implementation and its interface (for example, there is no as_slice method - the as_slices method returns a pair of slices) and has negative performance consequences (e.g. need to account for wrap around while iterating over the elements).

This crates provides SliceDeque, a double-ended queue implemented with a growable virtual ring-buffer.

A virtual ring-buffer implementation is very similar to the one used in VecDeque. The main difference is that a virtual ring-buffer maps two adjacent regions of virtual memory to the same region of physical memory:

// Virtual memory:
//
//  __________region_0_________ __________region_1_________
// [ 0 | 0 | 0 | T | T | T | 0 | 0 | 0 | 0 | T | T | T | 0 ]
//               ^:head  ^:tail
//
// Physical memory:
//
// [ 0 | 0 | 0 | T | T | T | 0 ]
//               ^:head  ^:tail

That is, both the virtual memory regions 0 and 1 above (top) map to the same physical memory (bottom). Just like VecDeque, when the queue grows beyond the end of the allocated physical memory region, the queue wraps around, and new elements continue to be appended at the beginning of the queue. However, because SliceDeque maps the physical memory to two adjacent memory regions, in virtual memory space the queue maintais the ilusion of a contiguous memory layout:

// Virtual memory:
//
//  __________region_0_________ __________region_1_________
// [ T | T | 0 | T | T | T | T | T | T | 0 | T | T | T | T ]
//               ^:head              ^:tail
//
// Physical memory:
//
// [ T | T | 0 | T | T | T | T ]
//       ^:tail  ^:head

Since processes in many Operating Systems only deal with virtual memory addresses, leaving the mapping to physical memory to the CPU Memory Management Unit (MMU), SliceDeque is able to Derefs into a slice in those systems.

This simplifies SliceDeque's API and implementation, giving it a performance advantage over VecDeque in some situations.

In general, you can think of SliceDeque as a Vec with O(1) pop_front and amortized O(1) push_front methods.

License

This project is licensed under either of

at your option.

Contribution

Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in SliceDeque by you, as defined in the Apache-2.0 license, shall be dual licensed as above, without any additional terms or conditions.