Difference between revisions of "Winter 2017 SPO600 Weekly Schedule"

From CDOT Wiki
Jump to: navigation, search
(Week 1 Deliverables)
(Evaluation)
 
(36 intermediate revisions by the same user not shown)
Line 20: Line 20:
 
|-
 
|-
  
|3||Jan 23||[[#Tuesday (Jan 26)|Assembly Lab (Lab 3) - continued]]||[[#Friday (Jan 29)|Compiled C Lab (Lab 4)]]||[[#Week 3 Deliverables|Blog about the  Assembly Lab (Lab 3) and Compiled C Lab (Lab 4)]]
+
|3||Jan 23||[[#Week 3 - Class I|Assembly Lab (Lab 3) - continued]]||[[#Week 3 - Class II|Compiled C Lab (Lab 4)]]||[[#Week 3 Deliverables|Blog about the  Assembly Lab (Lab 3) and Compiled C Lab (Lab 4)]]
 
|-
 
|-
  
|4||Jan 30||[[#Tuesday (Feb 2)|Software Optimization]]||[[#Friday (Feb 5)|Algorithm Selection Lab (Lab 5)]]||[[#Week 4 Deliverables|Blog about the Algorithm Selection Lab (Lab 5).]]
+
|4||Jan 30||[[#Week 4 - Class I|Software Optimization]]||[[#Week 4 - Class II|Algorithm Selection Lab (Lab 5)]]||[[#Week 4 Deliverables|Blog about the Algorithm Selection Lab (Lab 5).]]
 
|-
 
|-
  
|5||Feb 6|||[[#Tuesday (Feb 9)|Algorithm Selection Lab (Lab 5) Continued]]||[[#Friday (Feb 12)|SIMD and Auto-Vectorization (Lab 6)]]||[[#Week 5 Deliverables|Blog your the Algorithm Selection Lab (Lab 5) and the Auto-Vectorization Lab (Lab 6).]]
+
|5||Feb 6|||[[#Week 5 - Class I|Algorithm Selection Lab (Lab 5) Continued]]||[[#Week 5 - Class II|SIMD and Auto-Vectorization (Lab 6)]]||[[#Week 5 Deliverables|Blog your the Algorithm Selection Lab (Lab 5) and the Auto-Vectorization Lab (Lab 6).]]
 
|-
 
|-
  
|6||Feb 13||[[#Tuesday (Feb 16)|Memory architecture]]||[[#Friday (Feb 19)|Inline Assembler Lab (Lab 7)]]||[[#Week 6 Deliverables|Blog about your Inline Assembler Lab (Lab 7).]]
+
|6||Feb 13||[[#Week 6 - Class I|Inline Assembler Lab (Lab 7)]]||[[#Week 6 - Class II|Inline Assembler Lab (Lab 7) continued...]]||[[#Week 6 Deliverables|Blog about your Inline Assembler Lab (Lab 7).]]
 
|-
 
|-
  
|7||Feb 20||[[#Tuesday (Feb 23)|Course Presentation Assignment]]||[[#Friday (Feb 26)|Project Startup]]||[[#Week 7 Deliverables|Blog about your selected presentation and project topics.]]
+
|7||Feb 20||[[#Week 7 - Class I|Memory Architecture]]||[[#Week 7 - Class II|Project Startup]]||[[#Week 7 Deliverables|Blog about your selected presentation and project topics.]]
  
 
|-style="background: #f0f0ff"
 
|-style="background: #f0f0ff"
Line 38: Line 38:
 
|-
 
|-
  
|8||Mar 6||style="background:#f0f0ff" colspan="2" align="center"|Linaro Connect - No classes. Prepare for your presentation and work on your project.||[[#Week 8 Deliverables|Prepare for your presentation and work on your project.]]
+
|8||Mar 6||[[#Week 8 - Class I|Project Discussion]]||[[#Week 8 - Class II|Presentations]]||[[#Week 8 Deliverables|Blog about your project.]]
 
|-
 
|-
  
|9||Mar 13||[[#Tuesday (Mar 15)|Presentations]]||[[#Friday (Mar 18)|Presentations]]||[[#Week 9 Deliverables|Blog about your Presentation.]]
+
|9||Mar 13||[[#Week 9 - Class I|Profiling]]||[[#Week 9 - Class II|Benchmarking]]||[[#Week 9 Deliverables|Blog about your project work.]]
 
|-
 
|-
  
Line 70: Line 70:
 
|Labs||align="right"|10%||See deliverables column above. All labs must be submitted by April 21.
 
|Labs||align="right"|10%||See deliverables column above. All labs must be submitted by April 21.
 
|-
 
|-
|Project work||align="right"|60%||3 stages: 15% (TBA) / 20% (TBA) / 25% (TBA)
+
|Project work||align="right"|60%||3 stages: 15% (Sunday, March 26) / 20% (Thursday, April 9) / 25% (Saturday, April 22)
 
|}
 
|}
  
Line 182: Line 182:
  
 
=== Week 2 Deliverables ===
 
=== Week 2 Deliverables ===
 +
 
* Blog your conclusion to the [[SPO600 Code Review Lab|Code Review Lab]] (Lab 1)
 
* Blog your conclusion to the [[SPO600 Code Review Lab|Code Review Lab]] (Lab 1)
  
 +
== Week 3 ==
 +
 +
=== Week 3 - Class I ===
 +
 +
* Continue group work on [[SPO600 Assembler Lab|Lab 3]].
 +
 +
=== Week 3 - Class II ===
 +
 +
* [[SPO600 Compiled C Lab]] (Lab 4)
 +
 +
=== Week 3 Deliverables ===
 +
 +
* Blog your conclusion to:
 +
** [[SPO600 Assembler Lab|Lab 3]]
 +
** [[SPO600 Compiled C Lab|Lab 4]]
 +
 +
== Week 4 ==
 +
 +
=== Week 4 - Class I ===
 +
 +
Software Optimization
 +
* [[Compiler Optimizations]]
 +
* [[Profile Guided Optimization]]
 +
* Algorithm Selection
 +
 +
=== Week 4 - Class II ===
 +
 +
* [[SPO600 Algorithm Selection Lab]] (Lab 5)
 +
 +
=== Week 4 Deliverables ===
 +
 +
* Blog about your Lab 5 results.
 +
 +
== Week 5 ==
 +
 +
=== Week 5 - Class I ===
 +
 +
* Finish the [[SPO600 Algorithm Selection Lab|Algorithm Selection Lab]]
 +
 +
=== Week 5 - Class II ===
 +
 +
* Introduction to Vector Processing/SIMD
 +
* [[SPO600 Vectorization Lab|Vectorization Lab]] (Lab 6)
 +
 +
=== Week 5 Deliverables ===
 +
 +
* Blog your results for the [[SPO600 Algorithm Selection Lab|Algorithm Selection Lab]] (Lab 5)
 +
* Blog your results for the [[SPO600 Vectorization Lab|Vectorization Lab]] (Lab 6)
 +
* For each of the above, be sure to include links to your code, detailed results, and your reflection on the lab.
 +
 +
== Week 6 ==
 +
 +
=== Week 6 - Class I ===
 +
* [[Inline Assembly Language]] -- often used for:
 +
*# Implementing a memory barrier
 +
*# Performing an [[Atomic Operation]]
 +
*#* '''Atomics''' are operations which must be completed in a single step (or appear to be completed in a single step) without potential interruption.
 +
*#* Wikipedia has a good basic overview of the need for atomicity in the article on [http://en.wikipedia.org/wiki/Linearizability Linerarizability]
 +
*# Gaining performance (by accessing processor features not exposed by the high-level language being used (C, C++, ...))
 +
* [[SPO600 Inline Assembler Lab|Inline Assembler Lab]] (Lab 7)
 +
 +
=== Week 6 - Class II ===
 +
* [[SPO600 Inline Assembler Lab|Inline Assembler Lab]] (Lab 7) continued...
 +
 +
=== Week 6 Deliverables ===
 +
* Blog your Lab 7 results.
 +
 +
== Week 7 ==
 +
 +
=== Week 7 - Class I ===
 +
 +
==== Overview/Review of Processor Operation ====
 +
 +
* Fetch-decode-dispatch-execute cycle
 +
* Pipelining
 +
* Branch Prediction
 +
* In-order vs. Out-of-order execution
 +
** Micro-ops
 +
 +
==== Memory Basics ====
 +
 +
* Organization of Memory
 +
** System organization
 +
** Process organization
 +
*** Text, data
 +
*** Stack
 +
*** Heap
 +
* Memory Speeds
 +
* Cache
 +
** Cache lookup
 +
** Cache synchronization and invalidation
 +
** Cache line size
 +
* Prefetch
 +
** Prefetch hinting
 +
 +
==== Memory Architecture ====
 +
 +
* Virtual Memory and Memory Management Units (MMUs)
 +
** General principles of VM and operation of MMUs
 +
** Memory protection
 +
*** Unmapped Regions
 +
*** Write Protection
 +
*** Execute Protection
 +
*** Privilege Levels
 +
** Swapping
 +
** Text sharing
 +
** Data sharing
 +
** Shared memory for Inter-Process Communication
 +
** Copy-on-Write (CoW)
 +
** Demand Loading
 +
** Memory mapped files
 +
 +
==== Memory Barriers ====
 +
'''Memory Barriers''' ensure that memory accesses are sequenced so that multiple threads, processes, cores, or IO devices see a predictable view of memory.
 +
* Leif Lindholm provides an excellent explanation of memory barriers.
 +
** Blog series - I recommend this series, especially the introduction, as a very clear explanation of memory barrier issues.
 +
*** Part 1 - [http://community.arm.com/groups/processors/blog/2011/03/22/memory-access-ordering--an-introduction Memory Access Ordering - An Introduction]
 +
*** Part 2 - [http://community.arm.com/groups/processors/blog/2011/04/11/memory-access-ordering-part-2--barriers-and-the-linux-kernel Memory Access Ordering Part 2 - Barriers and the Linux Kernel]
 +
*** Part 3 - [http://community.arm.com/groups/processors/blog/2011/10/19/memory-access-ordering-part-3--memory-access-ordering-in-the-arm-architecture Memory Access Ordering Part 3 - Memory Access Ordering in the ARM Architecture]
 +
** Presentation at Embedded Linux Conference 2010 (Note: Acquire/Release in C++11 and ARMv8 aarch64 appeared after this presentation):
 +
*** [http://elinux.org/images/f/fa/Software_implications_memory_systems.pdf Slides]
 +
*** [http://free-electrons.com/pub/video/2010/elce/elce2010-lindholm-memory-450p.webm Video]
 +
* [http://www.rdrop.com/users/paulmck/scalability/paper/whymb.2010.07.23a.pdf Memory Barriers - A Hardware View for Software Hackers] - This is a highly-rated paper that explains memory barrier issues - as the title suggests, it is designed to describe the hardware origin of the problem to software developers. Despite the fact that it is an introduction to the topic, it is still very technical.
 +
* [http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.faqs/ka14041.html ARM Technical Support Knowlege Article - In what situations might I need to insert memory barrier instructions?] - Note that there are some additional mechanisms present in ARMv8 aarch64, including Acquire/Release.
 +
* [https://www.kernel.org/doc/Documentation/memory-barriers.txt Kernel Documentation on Memory Barriers] - discusses the memory barrier issue generally, and the solutions used within the Linux kernel. This is part of the kernel documentation.
 +
* Acquire-Release mechanisms
 +
** [http://blogs.msdn.com/b/oldnewthing/archive/2008/10/03/8969397.aspx MSDN Blog Post] with a very clear explanation of Acquire-Release.
 +
** [http://preshing.com/20130922/acquire-and-release-fences/ Preshing on Programming post] with a good explanation.
 +
** [http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.genc010197a/index.html ARMv8 Instruction Set Architecture Manual] (ARM InfoCentre registration required) - See the section on Acquire/Release and Load/Store, especially Load/Store Exclusive (e.g., LDREX)
 +
 +
==== The Future of Memory ====
 +
 +
* NUMA (on steroids!)
 +
* Non-volatile, byte-addressed main memory
 +
* Non-local memory
 +
* Memory encryption
 +
 +
=== Week 7 - Class II ===
 +
 +
* [[Winter 2017 SPO600 Project|Course Project]]
 +
 +
=== Week 7 Deliverables ===
 +
 +
* Blog your Lab 7 results, including the second part
 +
* (To be announced: Project Deliverables)
 +
 +
== Week 8 ==
 +
 +
=== Week 8 - Class I ===
 +
 +
* Project Discussions
 +
 +
=== Week 8 - Class II ===
 +
 +
* Project Presentation #0
 +
** Selected glibc function(s)
 +
** Plan of Action
 +
 +
=== Week 8 Deliverables ===
 +
 +
* Blog about your selected function(s) and project plan
 +
** Remember: You should be posting 1-2 times per week
 
<!--
 
<!--
 +
 +
#################################################################################
 +
#################################################################################
 +
#################################################################################
 +
 
== Week 3 ==
 
== Week 3 ==
  
Line 232: Line 400:
 
* [[SPO600 Vectorization Lab|Vectorization Lab]] (Lab 6)
 
* [[SPO600 Vectorization Lab|Vectorization Lab]] (Lab 6)
  
==== Week 5 Deliverables ====
+
=== Week 5 Deliverables ===
  
 
* Blog your results for the [[SPO600 Algorithm Selection Lab|Algorithm Selection Lab]] (Lab 5)
 
* Blog your results for the [[SPO600 Algorithm Selection Lab|Algorithm Selection Lab]] (Lab 5)

Latest revision as of 14:23, 4 April 2017

This is the schedule and main index page for the SPO600 Software Portability and Optimization course for Winter 2017.

Important.png
It's Alive!
This SPO600 weekly schedule will be updated as the course proceeds - dates and content are subject to change. The cells in the summary table will be linked to relevant resources and labs as the course progresses.

Schedule Summary Table

This is a summary/index table. Please follow the links in each cell for additional detail which will be added below as the course proceeds -- especially for the Deliverables column.

Week Week of... Class I
Tuesday 1:30-3:15
Class II
Friday 1:30-3:15
Deliverables
(Summary - click for details)
1 Jan 9 Introduction to Software Porting, Portability, Benchmarking, and Optimization / How is code accepted into an open source project? (Lab 1) Overview of Working with Code and Building Software (Lab 2) Set up accounts, and blog your conclusion to Lab 2.
2 Jan 16 Computer Architecture Overview Assembly Lab (Lab 3) Blog about the Code Review Lab (Lab 1)
3 Jan 23 Assembly Lab (Lab 3) - continued Compiled C Lab (Lab 4) Blog about the Assembly Lab (Lab 3) and Compiled C Lab (Lab 4)
4 Jan 30 Software Optimization Algorithm Selection Lab (Lab 5) Blog about the Algorithm Selection Lab (Lab 5).
5 Feb 6 Algorithm Selection Lab (Lab 5) Continued SIMD and Auto-Vectorization (Lab 6) Blog your the Algorithm Selection Lab (Lab 5) and the Auto-Vectorization Lab (Lab 6).
6 Feb 13 Inline Assembler Lab (Lab 7) Inline Assembler Lab (Lab 7) continued... Blog about your Inline Assembler Lab (Lab 7).
7 Feb 20 Memory Architecture Project Startup Blog about your selected presentation and project topics.
Study Week Feb 27 Study Week - No classes!
8 Mar 6 Project Discussion Presentations Blog about your project.
9 Mar 13 Profiling Benchmarking Blog about your project work.
10 Mar 20 Project Stage I Updates Discussion & Hack Session Blog about your Stage I Update.
11 Mar 27 Discussion & Hack Session Discussion & Hack Session Blog about your project work.
12 Apr 3 Discussion & Hack Session Project Stage II Updates Blog about your Stage II Update.
13 Apr 10 Wrap-Up Discussion Project Stage III Updates Blog about your project, including the Stage III Update, and write a wrap-up post about the course.
Exam Week Apr 18 Exam Week - No exam in this course!

Evaluation

Category Percentage Evaluation Dates
Communication 20% 5% each: End of January, end of February, end of March, end of course.
Quizzes 10% May be held during any class, usually at the start of class. A minimum of 5 one-page quizzes will be given. No make-up/retake option is offered if you miss a quiz. Lowest 3 scores will not be counted.
Labs 10% See deliverables column above. All labs must be submitted by April 21.
Project work 60% 3 stages: 15% (Sunday, March 26) / 20% (Thursday, April 9) / 25% (Saturday, April 22)

Week 1

Week 1 - Class I

Introduction to the Problems

Porting and Portability
  • Most software is written in a high-level language which can be compiled into machine code for a specific computer architecture. In many cases, this code can be compiled for multiple architectures. However, there is a lot of existing code that contains some architecture-specific code fragments written in Assembly Language (or, in some cases, machine-specific high-level code).
  • Reasons for writing code in Assembly Langauge include:
    • Performance
    • Atomic Operations
    • Direct access to hardware features, e.g., CPUID registers
  • Most of the historical reasons for including assembler are no longer valid. Modern compilers can out-perform most hand-optimized assembly code, atomic operations can be handled by libraries or compiler intrinsics, and most hardware access should be performed through the operating system or appropriate libraries.
  • A new architecture has appeared: AArch64, which is part of ARMv8. This is the first new computer architecture to appear in several years (at least, the first mainstream computer architecture).
  • At this point, most key open source software (the software typically present in a Linux distribution such as Ubuntu or Fedora, for example) now runs on AArch64. However, it may not run as well as on older architectures (such as x86_64).
Benchmarking and Profiling

Benchmarking involves testing software performance under controlled conditions so that the performance can be compared to other software, the same software operating on other types of computers, or so that the impact of a change to the software can be gauged.

Profiling is the process of analyzing software performance on finer scale, determining resource usage per program part (typically per function/method). This can identify software bottlenecks and potential targets for optimization.

Optimization

Optimization is the process of evaluating different ways that software can be written or built and selecting the option that has the best performance tradeoffs.

Optimization may involve substituting software algorithms, altering the sequence of operations, using architecture-specific code, or altering the build process. It is important to ensure that the optimized software produces correct results and does not cause an unacceptable performance regression for other use-cases, system configurations, operating systems, or architectures.

The definition of "performance" varies according to the target system and the operating goals. For example, in some contexts, low memory or storage usage is important; in other cases, fast operation; and in other cases, low CPU utilization or long battery life may be the most important factor. It is often possible to trade off performance in one area for another; using a lookup table, for example, can reduce CPU utilization and improve battery life in some algorithms, in return for increased memory consumption.

Most advanced compilers perform some level of optimization, and the options selected for compilation can have a significant effect on the trade-offs made by the compiler, affecting memory usage, execution speed, executable size, power consumption, and debuggability.

Build Process

Building software is a complex task that many developers gloss over. The simple act of compiling a program invokes a process with five or more stages, including pre-proccessing, compiling, optimizing, assembling, and linking. However, a complex software system will have hundreds or even thousands of source files, as well as dozens or hundreds of build configuration options, auto configuration scripts (cmake, autotools), build scripts (such as Makefiles) to coordinate the process, test suites, and more.

The build process varies significantly between software packages. Most software distribution projects (including Linux distributions such as Ubuntu and Fedora) use a packaging system that further wraps the build process in a standardized script format, so that different software packages can be built using a consistent process.

In order to get consistent and comparable benchmark results, you need to ensure that the software is being built in a consistent way. Altering the build process is one way of optimizing software.

Note that the build time for a complex package can range up to hours or even days!

General Course Information

  • Course resources are linked from the CDOT wiki, starting at http://zenit.senecac.on.ca/wiki/index.php/SPO600 (Quick find: This page will usually be Google's top result for a search on "SPO600").
  • Coursework is submitted by blogging.
  • Quizzes will be short (1 page) and will be held without announcement at any time. Your lowest three quiz scores will not be counted, so do not worry if you miss one or two.
  • Course marks (see Weekly Schedule for dates):
    • 60% - Project Deliverables
    • 20% - Communication (Blog and Wiki writing)
    • 20% - Labs and Quizzes (10% labs - completed/not completed; 10% for quizzes - lowest 3 scores not counted)
  • All classes will be held in an Active Learning Classroom -- you are encouraged to bring your own laptop to class. If you do not have a laptop, consider signing one out of the Learning Commons for class, or using a smartphone with an HDMI adapter.
  • For more course information, refer to the SPO600 Weekly Schedule (this page), the Course Outline, and SPO600 Course Policies.
  • Optional: You can participate in the Linaro Code Porting/Optimization contest. For details, see the YouTube video of Jon "maddog" Hall and Steve Mcintyre at Linaro Connect USA 2013.

Discussion of how open source communities work

Week 1 - Class II

  • Overview of the Build and Release Process
  • Working with Code
    1. Getting Code
      • In a tarball
      • From git
        • Git basics
      • Working with other version control systems
    2. Getting and Installing Build Dependencies
      • Required tools
      • Required libraries, headers, and modules
    3. Building the Code
      • Configuration tools (autotools, cmake)
      • Make
      • The compiler toolchain
        • Preprocessor
        • Compiler
        • Assembler
        • Linker
      • Debug vs. Non-debug/Stripped binaries
      • Installation Scripts
  • Looking at How Distributions Package the Code
    • Using fedpkg

Week 1 Deliverables

  1. Course setup:
    1. Set up your SPO600 Communication Tools - in particular, set up a blog and add it to Planet CDOT (via the Planet CDOT Feed List).
    2. Add yourself to the Winter 2017 SPO600 Participants page (leave the projects columns blank).
    3. Generate a pair of keys for SSH and email the public key to your professor.
    4. Sign and return the Open Source Professional Option Student Agreement.
  2. Complete Labs
    1. Code Review Lab (Lab 1) (Due end of week 2)
    2. Code Building Lab (Lab 2) (Due end of week 2)
  3. Optional (recommended): Set up a personal Fedora system.
  4. Optional: Purchase an AArch64 development board (such as a 96Boards HiKey or Raspberry Pi 3).

Week 2

Week 2 - Class I

Week 2 - Class II

Week 2 Deliverables

Week 3

Week 3 - Class I

  • Continue group work on Lab 3.

Week 3 - Class II

Week 3 Deliverables

Week 4

Week 4 - Class I

Software Optimization

Week 4 - Class II

Week 4 Deliverables

  • Blog about your Lab 5 results.

Week 5

Week 5 - Class I

Week 5 - Class II

Week 5 Deliverables

  • Blog your results for the Algorithm Selection Lab (Lab 5)
  • Blog your results for the Vectorization Lab (Lab 6)
  • For each of the above, be sure to include links to your code, detailed results, and your reflection on the lab.

Week 6

Week 6 - Class I

  • Inline Assembly Language -- often used for:
    1. Implementing a memory barrier
    2. Performing an Atomic Operation
      • Atomics are operations which must be completed in a single step (or appear to be completed in a single step) without potential interruption.
      • Wikipedia has a good basic overview of the need for atomicity in the article on Linerarizability
    3. Gaining performance (by accessing processor features not exposed by the high-level language being used (C, C++, ...))
  • Inline Assembler Lab (Lab 7)

Week 6 - Class II

Week 6 Deliverables

  • Blog your Lab 7 results.

Week 7

Week 7 - Class I

Overview/Review of Processor Operation

  • Fetch-decode-dispatch-execute cycle
  • Pipelining
  • Branch Prediction
  • In-order vs. Out-of-order execution
    • Micro-ops

Memory Basics

  • Organization of Memory
    • System organization
    • Process organization
      • Text, data
      • Stack
      • Heap
  • Memory Speeds
  • Cache
    • Cache lookup
    • Cache synchronization and invalidation
    • Cache line size
  • Prefetch
    • Prefetch hinting

Memory Architecture

  • Virtual Memory and Memory Management Units (MMUs)
    • General principles of VM and operation of MMUs
    • Memory protection
      • Unmapped Regions
      • Write Protection
      • Execute Protection
      • Privilege Levels
    • Swapping
    • Text sharing
    • Data sharing
    • Shared memory for Inter-Process Communication
    • Copy-on-Write (CoW)
    • Demand Loading
    • Memory mapped files

Memory Barriers

Memory Barriers ensure that memory accesses are sequenced so that multiple threads, processes, cores, or IO devices see a predictable view of memory.

The Future of Memory

  • NUMA (on steroids!)
  • Non-volatile, byte-addressed main memory
  • Non-local memory
  • Memory encryption

Week 7 - Class II

Week 7 Deliverables

  • Blog your Lab 7 results, including the second part
  • (To be announced: Project Deliverables)

Week 8

Week 8 - Class I

  • Project Discussions

Week 8 - Class II

  • Project Presentation #0
    • Selected glibc function(s)
    • Plan of Action

Week 8 Deliverables

  • Blog about your selected function(s) and project plan
    • Remember: You should be posting 1-2 times per week