2010-06-27 10:58:44 +04:00
name: speculation
2010-06-28 03:47:02 +04:00
version: 0.7.0
2010-06-27 10:58:44 +04:00
license: BSD3
license-file: LICENSE
author: Edward A. Kmett
maintainer: Edward A. Kmett <ekmett@gmail.com>
stability: experimental
2010-06-27 11:44:55 +04:00
homepage: http://github.com/ekmett/speculation
2010-06-27 10:58:44 +04:00
category: Concurrency
2010-06-27 11:14:35 +04:00
synopsis: A framework for safe, programmable, speculative parallelism
2010-06-27 12:26:28 +04:00
description:
2010-06-28 02:30:18 +04:00
A framework for safe, programmable, speculative parallelism, loosely based on:
2010-06-27 19:02:55 +04:00
.
2010-06-28 02:30:18 +04:00
* Prakash Prabhu, G. Ramalingam, and Kapil Vaswani, \"/Safe Programmable Speculative Parallelism/\",
In the proceedings of Programming Language Design and Implementation (PLDI) Vol 45, Issue 6 (June 2010) pp 50-61.
<http://research.microsoft.com/pubs/118795/pldi026-vaswani.pdf>
.
This package provides speculative function application and speculative folds. Speculative STM transactions take the place
2010-06-27 19:02:55 +04:00
of the transactional rollback machinery from the paper.
.
For example:
.
2010-06-28 02:02:59 +04:00
@'spec' g f a@ evaluates @f g@ while forcing @a@, if @g == a@ then @f g@ is returned, otherwise @f a@ is evaluated and returned. Furthermore, if the argument has already been evaluated, we skip the @f g@ computation entirely. If a good guess at the value of @a@ is available, this is one way to induce parallelism in an otherwise sequential task. However, if the guess isn\'t available more cheaply than the actual answer, then this saves no work and if the guess is wrong, you risk evaluating the function twice.
2010-06-27 19:02:55 +04:00
.
The best-case timeline looks like:
.
> [---- f g ----]
> [----- a -----]
> [-- spec g f a --]
.
The worst-case timeline looks like:
.
> [---- f g ----]
> [----- a -----]
> [---- f a ----]
> [------- spec g f a -----------]
.
Compare these to the timeline of @f $! a@:
.
> [---- a -----]
> [---- f a ----]
.
'specSTM' provides a similar time table for STM actions, but also rolls back side-effects.
.
2010-06-28 03:47:02 +04:00
/Changes in 0.7.0:/
2010-06-28 03:26:42 +04:00
.
2010-06-28 03:40:53 +04:00
* Changed @'throw' 'SpeculationException'@ to 'retry'
2010-06-28 03:47:02 +04:00
* Removed 'SpeculationException'
2010-06-28 03:26:42 +04:00
.
2010-06-28 02:57:10 +04:00
/Changes in 0.6.0:/
.
2010-06-28 03:47:02 +04:00
* Upgraded the comparisons used by the STM combinators to STM actions, so they can check other STM state
2010-06-28 02:57:10 +04:00
.
2010-06-28 02:02:59 +04:00
/Changes in 0.5.1:/
2010-06-27 19:02:55 +04:00
.
2010-06-28 03:40:53 +04:00
* Exposed 'unsafeGetTagBits' and 'unsafeIsEvaluated'
2010-06-27 12:20:12 +04:00
2010-06-27 19:02:55 +04:00
copyright: (c) 2010 Edward A. Kmett
build-type: Simple
cabal-version: >=1.2
tested-with: GHC==6.12.1
2010-06-27 11:14:35 +04:00
extra-source-files: README.markdown
2010-06-27 10:58:44 +04:00
library
ghc-options: -Wall
build-depends:
2010-06-27 12:26:28 +04:00
base >= 4 && < 6,
parallel >= 2.2 && < 2.3,
stm >= 2.1 && < 2.2
2010-06-27 10:58:44 +04:00
exposed-modules:
Control.Concurrent.Speculation
2010-06-27 19:02:55 +04:00
Data.Foldable.Speculation