1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492
// Copyright 2015 Linus Färnstrand // // Licensed under the Apache License, Version 2.0 (the "License"); // you may not use this file except in compliance with the License. // You may obtain a copy of the License at // // http://www.apache.org/licenses/LICENSE-2.0 // // Unless required by applicable law or agreed to in writing, software // distributed under the License is distributed on an "AS IS" BASIS, // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. // See the License for the specific language governing permissions and // limitations under the License. //! # ForkJoin //! A work stealing fork-join parallelism library. //! //! [![Build Status](https://api.travis-ci.org/faern/forkjoin.svg?branch=master)](https://travis-ci.org/faern/forkjoin) //! //! Inspired by the blog post [Data Parallelism in Rust](http://smallcultfollowing.com/babysteps/blog/2013/06/11/data-parallelism-in-rust/) //! and implemented as part of a master's thesis. Repository hosted at [github.com/faern/forkjoin](https://github.com/faern/forkjoin) //! //! Library documentation hosted [here](https://faern.github.io/rust-docs/forkjoin/forkjoin/) //! //! This library has been developed to accommodate the needs of three types of //! algorithms that all fit very well for fork-join parallelism. //! //! # Reduce style //! //! Reduce style is where the algorithm receive an argument, recursively compute a value //! from this argument and return one answer. Examples of this style include recursively //! finding the n:th Fibonacci number and summing of tree structures. //! Characteristics of this style is that the algorithm does not need to mutate its //! argument and the resulting value is only available after every subtask has been //! fully computed. //! //! In reduce style algorithms the return values of each subtask is passed to a special //! join function that is executed when all subtasks have completed. //! To this join function an extra argument can be sent directly from the task if the algorithm //! has has `ReduceStyle::Arg`. This can be seen in the examples here. //! //! ## Example of reduce style (`ReduceStyle::NoArg`) //! //! ```rust //! use forkjoin::{TaskResult,ForkPool,AlgoStyle,ReduceStyle,Algorithm}; //! //! fn fib_30_with_4_threads() { //! let forkpool = ForkPool::with_threads(4); //! let fibpool = forkpool.init_algorithm(Algorithm { //! fun: fib_task, //! style: AlgoStyle::Reduce(ReduceStyle::NoArg(fib_join)), //! }); //! //! let job = fibpool.schedule(30); //! let result: usize = job.recv().unwrap(); //! assert_eq!(1346269, result); //! } //! //! fn fib_task(n: usize, _: usize) -> TaskResult<usize, usize> { //! if n < 2 { //! TaskResult::Done(1) //! } else { //! TaskResult::Fork(vec![n-1,n-2], None) //! } //! } //! //! fn fib_join(values: &[usize]) -> usize { //! values.iter().fold(0, |acc, &v| acc + v) //! } //! ``` //! //! ## Example of reduce style (`ReduceStyle::Arg`) //! //! ```rust //! use forkjoin::{TaskResult,ForkPool,AlgoStyle,ReduceStyle,Algorithm}; //! //! struct Tree { //! value: usize, //! children: Vec<Tree>, //! } //! //! fn sum_tree(t: &Tree) -> usize { //! let forkpool = ForkPool::new(); //! let sumpool = forkpool.init_algorithm(Algorithm { //! fun: sum_tree_task, //! style: AlgoStyle::Reduce(ReduceStyle::Arg(sum_tree_join)), //! }); //! let job = sumpool.schedule(t); //! job.recv().unwrap() //! } //! //! fn sum_tree_task(t: &Tree, _: usize) -> TaskResult<&Tree, usize> { //! if t.children.is_empty() { //! TaskResult::Done(t.value) //! } else { //! let mut fork_args: Vec<&Tree> = vec![]; //! for c in t.children.iter() { //! fork_args.push(c); //! } //! TaskResult::Fork(fork_args, Some(t.value)) // Pass current nodes value to join //! } //! } //! //! fn sum_tree_seq(t: &Tree) -> usize { //! t.value + t.children.iter().fold(0, |acc, t2| acc + sum_tree_seq(t2)) //! } //! //! fn sum_tree_join(value: &usize, values: &[usize]) -> usize { //! *value + values.iter().fold(0, |acc, &v| acc + v) //! } //! ``` //! //! # Search style //! //! Search style return results continuously and can sometimes start without any //! argument, or start with some initial state. The algorithm produce one or multiple //! output values during the execution, possibly aborting anywhere in the middle. //! Algorithms where leafs in the problem tree represent a complete solution to the //! problem (unless the leaf represent a dead end that is not a solution and does //! not spawn any subtasks), for example nqueens and sudoku solvers, have this style. //! Characteristics of the search style is that they can produce multiple results //! and can abort before all tasks in the tree have been computed. //! //! ## Example of search style //! //! ```rust //! use forkjoin::{ForkPool,TaskResult,AlgoStyle,Algorithm}; //! //! type Queen = usize; //! type Board = Vec<Queen>; //! type Solutions = Vec<Board>; //! //! fn search_nqueens() { //! let n: usize = 8; //! let empty = vec![]; //! //! let forkpool = ForkPool::with_threads(4); //! let queenpool = forkpool.init_algorithm(Algorithm { //! fun: nqueens_task, //! style: AlgoStyle::Search, //! }); //! //! let job = queenpool.schedule((empty, n)); //! //! let mut solutions: Vec<Board> = vec![]; //! loop { //! match job.recv() { //! Err(..) => break, // Job has completed //! Ok(board) => solutions.push(board), //! }; //! } //! let num_solutions = solutions.len(); //! println!("Found {} solutions to nqueens({}x{})", num_solutions, n, n); //! } //! //! fn nqueens_task((q, n): (Board, usize), _: usize) -> TaskResult<(Board,usize), Board> { //! if q.len() == n { //! TaskResult::Done(q) //! } else { //! let mut fork_args: Vec<(Board, usize)> = vec![]; //! for i in 0..n { //! let mut q2 = q.clone(); //! q2.push(i); //! //! if ok(&q2[..]) { //! fork_args.push((q2, n)); //! } //! } //! TaskResult::Fork(fork_args, None) //! } //! } //! //! fn ok(q: &[usize]) -> bool { //! for (x1, &y1) in q.iter().enumerate() { //! for (x2, &y2) in q.iter().enumerate() { //! if x2 > x1 { //! let xd = x2-x1; //! if y1 == y2 || y1 == y2 + xd || (y2 >= xd && y1 == y2 - xd) { //! return false; //! } //! } //! } //! } //! true //! } //! ``` //! //! # In-place mutation style //! //! NOTE: This style works in the current lib version, but it requires very ugly //! unsafe code! //! //! In-place mutation style receive a mutable argument, recursively modifies this value //! and the result is the argument itself. Sorting algorithms that sort their input //! arrays are cases of this style. Characteristics of this style is that they mutate //! their input argument instead of producing any output. //! //! Examples of this will come when they can be nicely implemented. //! //! # Tasks //! //! The small units that are executed and can choose to fork or to return a value is the //! `TaskFun`. A TaskFun can NEVER block, because that would block the kernel thread //! it's being executed on. Instead it should decide if it's done calculating or need //! to fork. This decision is taken in the return value to indicate to the user //! that a TaskFun need to return before anything can happen. //! //! A TaskFun return a `TaskResult`. It can be `TaskResult::Done(value)` if it's done //! calculating. It can be `TaskResult::Fork(args)` if it needs to fork. //! //! # TODO //! //! - [ ] Make mutation style algorithms work without giving join function //! - [ ] Implement a sorting algorithm. Quicksort? //! - [ ] Remove need to return None on fork with NoArg //! - [ ] Make it possible to use algorithms with different Arg & Ret on same pool. //! - [ ] Make ForkJoin work in stable Rust. //! - [ ] Remove mutex around channel in search style. #![feature(unique)] #![feature(scoped)] #![feature(libc)] extern crate deque; extern crate rand; extern crate num_cpus; extern crate libc; use std::ptr::Unique; use std::sync::atomic::AtomicUsize; use std::sync::{Arc,Mutex}; use std::sync::mpsc::{channel,Sender,Receiver,TryRecvError}; use std::fmt; use std::thread; mod workerthread; mod poolsupervisor; use ::poolsupervisor::{PoolSupervisorThread,SupervisorMsg}; /// Type definition of the main function in a task. /// Your task functions must have this signature pub type TaskFun<Arg, Ret> = extern "Rust" fn(Arg, usize) -> TaskResult<Arg, Ret>; /// Type definition of functions joining together forked results. /// Only used in `AlgoStyle::Reduce` algorithms with `ReduceStyle::NoArg`. pub type TaskJoin<Ret> = extern "Rust" fn(&[Ret]) -> Ret; /// Similar to `TaskJoin` but takes an extra argument sent directly /// from the task in algorithms with `ReduceStyle::Arg`. pub type TaskJoinArg<Ret> = extern "Rust" fn(&Ret, &[Ret]) -> Ret; /// Internal representation of a task. pub struct Task<Arg: Send, Ret: Send + Sync> { pub algo: Algorithm<Arg, Ret>, pub arg: Arg, pub join: ResultReceiver<Ret>, } /// Return values from tasks. Represent a computed value or a fork of the algorithm. pub enum TaskResult<Arg, Ret> { /// Return this from `TaskFun` to indicate a computed value. Represents a leaf in the /// problem tree of the computation. /// /// If the algorithm style is `AlgoStyle::Search` the value in `Done` will be sent /// directly to the `Job` held by the submitter of the computation. /// If the algorithm style is `AlgoStyle::Reduce` the value in `Done` will be inserted /// into a join barrier and later passed to the algorithms join function when all /// subtasks have completed execution. Done(Ret), /// Return this from `TaskFun` to indicate that the algorithm wants to fork. /// Takes a list of arguments to fork on. One subtask will be created for each argument. /// The second value is only used by `ReduceStyle::Arg`-algorithms to send a value directly /// to the `TaskJoinArg`, passing None in such algorithms will yield a panic. Fork(Vec<Arg>, Option<Ret>), } /// Enum representing the style of the executed algorithm. pub enum AlgoStyle<Ret> { /// A `Reduce` style algorithm join together the results of the individual nodes /// in the problem tree to finally form one result for the entire computation. /// /// Examples of this style include recursively computing fibbonacci numbers /// and summing binary trees. /// /// Takes a `ReduceStyle` to indicate what type of join function to use. Reduce(ReduceStyle<Ret>), /// A `Search` style algoritm return their results to the listener directly upon a /// `TaskResult::Done`. /// /// Examples of this style include sudoku solvers and nqueens where a node can /// represent a complete solution. Search, } impl<Ret> Copy for AlgoStyle<Ret> {} impl<Ret> Clone for AlgoStyle<Ret> { fn clone(&self) -> AlgoStyle<Ret> { *self } } /// Enum indicating what type of join function an `Algorithm` will use. pub enum ReduceStyle<Ret> { /// No extra argument is passed to the join function, only the resulting values of all subtasks NoArg(TaskJoin<Ret>), /// Together with the result values of the subtasks, the join function will also /// be passed an argument sent directly from the `TaskFun`. Arg(TaskJoinArg<Ret>), } impl<Ret> Copy for ReduceStyle<Ret> {} impl<Ret> Clone for ReduceStyle<Ret> { fn clone(&self) -> ReduceStyle<Ret> { *self } } /// The representation of a specific algorithm to use the ForkJoin library. /// /// Create one instance of this struct for each algorithm to be executed in ForkJoin. pub struct Algorithm<Arg: Send, Ret: Send + Sync> { /// A function pointer. The function that will be executed by all the tasks and subtasks. pub fun: TaskFun<Arg, Ret>, /// Enum showing the type of algorithm, indicate what should be done with results from /// subtasks created by forks of this algorithm. pub style: AlgoStyle<Ret>, } impl<Arg: Send, Ret: Send + Sync> Copy for Algorithm<Arg,Ret> {} impl<Arg: Send, Ret: Send + Sync> Clone for Algorithm<Arg,Ret> { fn clone(&self) -> Algorithm<Arg,Ret> { *self } } /// Internal struct for receiving results from multiple subtasks in parallel pub struct JoinBarrier<Ret: Send + Sync> { /// Atomic counter counting missing arguments before this join can be executed. pub ret_counter: AtomicUsize, /// Function to execute when all arguments have arrived. pub joinfun: ReduceStyle<Ret>, /// Extra argument to pass to `joinfun` only used when `joinfun` is `ReduceStyle::Arg`. pub joinarg: Option<Ret>, /// Vector holding the results of all subtasks. Initialized unsafely so can't be used /// for anything until all the values have been put in place. pub joinfunarg: Vec<Ret>, /// Where to send the result of the execution of `joinfun` pub parent: ResultReceiver<Ret>, } /// Enum describing what to do with results of `Task`s and `JoinBarrier`s. pub enum ResultReceiver<Ret: Send + Sync> { /// Algorithm has Reduce style and the value should be inserted into a `JoinBarrier` Join(Unique<Ret>, Arc<JoinBarrier<Ret>>), /// Algorithm has Search style and results should be sent directly to the owner. Channel(Arc<Mutex<Sender<Ret>>>), } impl<Ret: Send + Sync> Clone for ResultReceiver<Ret> { fn clone(&self) -> Self { match *self { ResultReceiver::Join(..) => panic!("Unable to clone ResultReceiver::Join"), ResultReceiver::Channel(ref c) => ResultReceiver::Channel(c.clone()), } } } /// Enum indicating there was a problem fetching a result from a job. #[derive(Debug)] pub enum ResultError { /// Returned from `try_recv` when no results are available at the time of the call NoResult, /// Returned from `try_recv` and `recv` when there are no more results to fetch Completed, } impl fmt::Display for ResultError { fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { let msg = match *self { ResultError::Completed => "The job has finished executing, no results left to read", ResultError::NoResult => "No results available", }; write!(f, "{}", msg) } } /// The handle for a computation. Can be used to fetch results of the computation. /// Upon drop it will wait for the entire computation to complete if it's still executing. /// Algorithm termination is detected by the `try_recv` and `recv` methods returning a `ResultError` pub struct Job<Ret> { port: Receiver<Ret>, } impl<Ret> Job<Ret> { /// Attempt to get a result from this `Job` without blocking. /// Will return a `ResultError` if no result is available at the time of call. pub fn try_recv(&self) -> Result<Ret, ResultError> { match self.port.try_recv() { Ok(res) => Ok(res), Err(e) => match e { TryRecvError::Empty => Err(ResultError::NoResult), TryRecvError::Disconnected => Err(ResultError::Completed), } } } /// Block and wait for a result. If a result is available it will return immediately. pub fn recv(&self) -> Result<Ret, ResultError> { match self.port.recv() { Ok(res) => Ok(res), Err(_) => Err(ResultError::Completed), } } } impl<Ret> Drop for Job<Ret> { /// Don't allow a job to be dropped while it's still computing. /// Block and fetch all results. fn drop(&mut self) { while let Ok(_) = self.port.recv() {} } } /// A handle for a specific `Algorithm` running on a `ForkPool`. /// Acquired from `ForkPool::init_algorithm`. pub struct AlgoOnPool<'a, Arg: 'a + Send, Ret: 'a + Send + Sync> { forkpool: &'a ForkPool<'a, Arg, Ret>, algo: Algorithm<Arg, Ret>, } impl<'a, Arg: Send, Ret: Send + Sync> AlgoOnPool<'a, Arg, Ret> { /// Schedule a new computation. Returns instantly with a handle to the computation. /// /// Return value(s) can be read from the returned `Job`. /// `AlgoStyle::Reduce` will only return one message on this channel. /// /// `AlgoStyle::Search` algorithm might return arbitrary number of messages. pub fn schedule(&self, arg: Arg) -> Job<Ret> { let (chan, port) = channel(); let task = Task { algo: self.algo, arg: arg, join: ResultReceiver::Channel(Arc::new(Mutex::new(chan))), }; self.forkpool.schedule(task); Job { port: port } } } /// Main struct of the ForkJoin library. /// Represents a pool of threads implementing a work stealing algorithm. pub struct ForkPool<'a, Arg: Send, Ret: Send + Sync> { #[allow(dead_code)] joinguard: thread::JoinGuard<'a, ()>, channel: Sender<SupervisorMsg<Arg, Ret>>, } impl<'a, Arg: Send + 'a, Ret: Send + Sync + 'a> ForkPool<'a, Arg, Ret> { /// Create a new `ForkPool` using num_cpus to determine pool size pub fn new() -> ForkPool<'a, Arg, Ret> { let nthreads = num_cpus::get(); ForkPool::with_threads(nthreads) } /// Create a new `ForkPool` with `nthreads` `WorkerThread`s at its disposal. pub fn with_threads(nthreads: usize) -> ForkPool<'a, Arg, Ret> { assert!(nthreads > 0); let (channel, joinguard) = PoolSupervisorThread::spawn(nthreads); ForkPool { joinguard: joinguard, channel: channel, } } /// Apply a specified algorithm to this `ForkPool` and get a handle for it where jobs /// can be scheduled. pub fn init_algorithm(&self, algorithm: Algorithm<Arg, Ret>) -> AlgoOnPool<Arg, Ret> { AlgoOnPool { forkpool: self, algo: algorithm, } } fn schedule(&self, task: Task<Arg, Ret>) { self.channel.send(SupervisorMsg::Schedule(task)).unwrap(); } } impl<'a, Arg: Send, Ret: Send + Sync> Drop for ForkPool<'a, Arg, Ret> { fn drop(&mut self) { match self.channel.send(SupervisorMsg::Shutdown) { Ok(_) => (), Err(e) => panic!("Unable to send Shutdown to supervisor: {}", e), } } }