Cell Programming IBM

Programming Frameworks

example

Currently there are two programming frameworks to use (we don't intend to talk about © Sony SDK nor about the free community driven psl1ght on this page). The Cell SDK is the most full featured and "supported" but is architecture specific, whereas the OpenCL should allow code portability to other accelerators (like GP-GPU's) however is from IBM alphaworks so should be considered beta.

• .! POSTED 06-2018 UNTESTED INTACT, Cell SDK some or most MARS packages, INTACT IBM C.BE MARS ps3.tar .gz POSTED 06-2018 http://mega.nz/#!TnR2QAiK!4nMgCZwRjCPe-JQUtehsurvo96DiUpt6EaXw5HCtDVc

Cell SDK 3.0

• CellSDK-Devel-Fedora_3.0.0.1.0.iso
• CellSDK-Extras-Fedora_3.0.0.1.0.iso
• SDK-Installation_Guide_v3.0.pdf
• v3.0 README Fedora 7
• additional stuff : SDK3.0 @ BSC.es

Cell SDK 3.1

• SDK 3.1 Installer : cell-install-3.1.0-0.0.noarch.rpm / mirror: cell-install-3.1.0-0.0.noarch.rpm (10.75 MB)

• SDK 3.1 Developer package ISO image for Fedora 9 : CellSDK-Devel-Fedora_3.1.0.0.0.iso (434MB) / mirror: CellSDK-Devel-Fedora_3.1.0.0.0.iso (433.65 MB)

• SDK 3.1 Extras package ISO image for Fedora 9 : CellSDK-Extras-Fedora_3.1.0.0.0.iso / mirror: CellSDK-Extras-Fedora_3.1.0.0.0.iso (33.36 MB)

• v3.1 README Fedora 9 / mirror: README.fedora9 (6.73 KB)

• additional stuff : SDK3.1 @ BSC.es

• Installation Quick Guide pastie

Cell SystemSim

note: The IBM Full-System Simulator has been removed from the SDK package, and is now available separately. To obtain the Simulator, download it from the following alphaWorks Web site: http://www.alphaworks.ibm.com/tech/cellsystemsim working latest mirror

• systemsim-cell-3.1-25.f9.i386.rpm / mirror: systemsim-cell-3.1-25.f9.i386.rpm (5.16 MB)

• systemsim-cell-3.1-25.f9.ppc64.rpm / mirror: systemsim-cell-3.1-25.f9.ppc64.rpm (5.93 MB)

• systemsim-cell-3.1-25.f9.x86_64.rpm / mirror: systemsim-cell-3.1-25.f9.x86_64.rpm (6.04 MB)

• README.txt / mirror: README.txt (5.55 KB)

• IBM Full System Simulator For The CBE Processor, version 3.1 (234.2 MB)

Compilers

• ppu-gcc -- GNU GCC PPC compiler

• spu-gcc -- GNU GCC SPU compiler

• xlcl -- IBM XL OpenCL compiler

Documentation

• Cell SDK
  • Programming:CBE Programmers Guide,CBE Handbook,CBE Programming Tutorial,CBE PXCell Handbook,Performance Tools Reference,IDE Users Guide
  • Libraries:3D FFT,libFFT,ALF,BLAS,SIMD Math API,DaCS,LAPACK,Monte Carlo,SDK Library Examples,SPE Runtime Management, SPE Runtime Migration,SPE Runtime Library Ext.
  • Architecture: CBEA,CBEA Public Registers,PPC Book 1,PPC Book 2,PPC Book 3,SPU ISA
  • Standards:CBE Linux ABI,Language Extentsions,SIMD Library Specification,SPU ABI Specification,SPU Assembly Language Specification
  • Installation: Installation Guide
• Programming the CellBE Architecture - Examples and Best Practices
• Cell OpenCL
  • OpenCL Guide,

Further Info

Even more docs http://moss.csc.ncsu.edu/~mueller/cluster/ps3/

Project OpenCLit - Cluster Computing with the PS3 : http://portal.gitbrew.org/wikibrew/PS3:OpenCLit

Tutorials

• Cell Programming Tutorial - IBM
  • part1 - Summary: Meet the Cell Broadband Engine™ (Cell BE) processor from a compiler-writer's perspective, and get a bird's-eye view of a number of the unique challenges it poses in this first tutorial of a five-part series.
  • part2 - Summary: This tutorial discusses specific issues in optimizing code to run effectively on the Synergistic Processor Elements (SPEs) in the Cell Broadband Engine™ (Cell BE) processor.
  • part3 - Summary: This tutorial discusses the compiler issues in optimizing code to run efficiently on SIMD-capable processors. In particular, it shows how to optimize code that must run both on the VMX SIMD engine of the PowerPC® core of the Cell Broadband Engine (Cell BE) processor, and also on the SIMD-only Synergistic Processor Elements (SPEs).
  • part4 - Summary: This part discusses ways to partition code to run across the multiple cores available in a Cell Broadband Engine™ (Cell BE) processor. It gives particular attention to efficient partitioning of code to allow larger programs or data sets to be manipulated using the 256KB of local store available on the Synergistic Processor Elements (SPEs).
  • part5 - Summary: This last tutorial discusses techniques for managing data in the local store of the Synergistic Processor Elements (SPEs) of a Cell Broadband Engine (Cell BE) processor. Learn particular techniques such as double-buffering and maintaining a reasonably efficient software cache.

• IBM Cell Broadband engine tutorials

• iLab University of Southern California 18 part course

• • 1.Intro ** 2.Cell ** 3.Architecture ** 4.Concurrency ** 5.Parallelism ** 6.Patterns1 ** 7.Patterns2 ** 8.Streamit ** 9.Debugging ** 10.Performance ** 11.Compilers ** 12.Streaming ** 13.Starp ** 14.Bluespec ** 15.Cilk ** 16.Games ** 17.Raw ** 18.Future

• PowerPC Architecture Book, Version 2.02
  • Book I: PowerPC User Instruction Set Architecture
  • Book II: PowerPC Virtual Environment Architecture
  • Book III: PowerPC Operating Environment Architecture

PowerPC Assembly

• Lightsoft: Beginners Guide to PowerPC Assembly Language

• Simplified Mnemonics
• Appendix F - Simplified Mnemonics

• SPU Instructions Visual Guide

Hypervisor

Research hypervisor (rhype):

• http://www.research.ibm.com/hypervisor/
• http://code.google.com/p/rhype/
• https://github.com/ajray/rhype (backup: ajray-rhype-fdb96b4.zip (1.04 MB)

Secure hypervisor(shype):

• http://www.research.ibm.com/secure_systems_department/projects/hypervisor/