Search icon CANCEL
Subscription
0
Cart icon
Your Cart (0 item)
Close icon
You have no products in your basket yet
Arrow left icon
Explore Products
Best Sellers
New Releases
Books
Videos
Audiobooks
Learning Hub
Free Learning
Arrow right icon
Arrow up icon
GO TO TOP
Haskell High Performance Programming

You're reading from   Haskell High Performance Programming Write Haskell programs that are robust and fast enough to stand up to the needs of today

Arrow left icon
Product type Paperback
Published in Sep 2016
Publisher Packt
ISBN-13 9781786464217
Length 408 pages
Edition 1st Edition
Languages
Tools
Arrow right icon
Author (1):
Arrow left icon
Samuli Thomasson Samuli Thomasson
Author Profile Icon Samuli Thomasson
Samuli Thomasson
Arrow right icon
View More author details
Toc

Table of Contents (16) Chapters Close

Preface 1. Identifying Bottlenecks FREE CHAPTER 2. Choosing the Correct Data Structures 3. Profile and Benchmark to Your Heart's Content 4. The Devil's in the Detail 5. Parallelize for Performance 6. I/O and Streaming 7. Concurrency and Performance 8. Tweaking the Compiler and Runtime System (GHC) 9. GHC Internals and Code Generation 10. Foreign Function Interface 11. Programming for the GPU with Accelerate 12. Scaling to the Cloud with Cloud Haskell 13. Functional Reactive Programming 14. Library Recommendations Index

Primitive GHC-specific features


All strictly GHC-specific functionality is contained in GHC.* module. The GHC.Exts module is of particular interest. The GHC.Prim module (re-exported by GHC.Exts) exports core primitives in GHC.

For a while now, GHC has shipped with primitives for SIMD processor instructions. These are available when compiling via the LLVM backend (-fllvm).

SIMD stands for Single Instruction, Multiple Data. It basically means performing the same operation on a whole vector of machine numbers at the cost of performing that operation on just one number. SIMD vector types can be found in the GHC.Prim module. The specialized vectors are named like Int8X16#, which stands for an Int8 vector of length 16. DoubleX8# stands for a vector of eight double precision values:

data Int8X16#
data DoubleX8#

These types are primitive and there are no exposed constructors.

To create vectors that can be used with SIMD instructions, we have two basic options. The first one is to use one of the broadcast...

lock icon The rest of the chapter is locked
Register for a free Packt account to unlock a world of extra content!
A free Packt account unlocks extra newsletters, articles, discounted offers, and much more. Start advancing your knowledge today.
Unlock this book and the full library FREE for 7 days
Get unlimited access to 7000+ expert-authored eBooks and videos courses covering every tech area you can think of
Renews at $19.99/month. Cancel anytime
Banner background image