Примож Габриэльчич - Высокая производительность Delphi (черновик перевода глав 1-2)

the reviewers

Packt is searching for authors like you

Preface

Who this book is for

What this book covers

To get the most out of this book

Download the example code files

Conventions used

Get in touch

Reviews

About Performance

What is performance?

Different types of speed

Algorithm complexity

Big O and Delphi data structures

Data structures in practice

Mr. Smith's first program

Looking at code through the Big O eyes

Don't guess, measure!

Profiling with TStopwatch

Profilers

AsmProfiler

Sampling Profiler

AQTime

Nexus Quality Suite

Summary

Fixing the Algorithm

Responsive user interfaces

Updating a progress bar

Bulk updates

Virtual display

Caching

Dynamic cache

Speeding up SlowCode

Summary

Fine-Tuning the Code

Delphi compiler settings

Code inlining control

Optimization

Record field alignment

Assertions

Overflow checking

Range checking

Extracting common expressions

The helpful CPU window

Behind the scenes

A plethora of types

Simple types

Strings

Arrays

Records

Classes

Interfaces

Optimizing method calls

Parameter passing

Method inlining

The magic of pointers

Going the assembler way

Returning to SlowCode

Summary

Memory Management

Optimizing strings and array allocations

Memory management functions

Dynamic record allocation

FastMM internals

Memory allocation in a parallel world

Replacing the default memory manager

ScaleMM

TBBMalloc

Fine-tuning SlowCode

Summary

Getting Started with the Parallel World

Processes and threads

When to parallelize the code?

Most common problems

Never access UI from a background thread

Simultaneous reading and writing

Sharing a variable

Synchronization

Critical sections

Other locking mechanisms

A short note on coding style

Shared data with built-in locking

Interlocked operations

Object life cycle

Communication

Windows messages

Synchronize and Queue

Polling

Performance

Third-party libraries

Summary

Working with Parallel Tools

TThread

Advanced TThread

Setting up a communication channel

Sending messages from a thread

Implementing a timer

Summary

Exploring Parallel Practices

Tasks and patterns

Variable capturing

Tasks

Exceptions in tasks

Parallelizing a loop

Thread pooling

Async/Await

Join

Join/Await

Future

Parallel for

Pipelines

Creating the pipeline

Stages

Displaying the result and shutting down

Summary

Using External Libraries

Using object files

Object file formats

Object file linking in practice

Using C++ libraries

Using a proxy DLL in Delphi

Summary

Best Practices

About performance

Fixing the algorithm

Fine-tuning the code

Memory management

Getting started with the parallel world

Working with parallel tools

Exploring parallel practices

Using external libraries

Final words

Other Books You May Enjoy

Leave a review - let other readers know what you think

Preface

Performance matters!

I started programming on 8-bit micros, and boy, was that an interesting time! Memory was typically not a problem as we didn't write big programs, but they certainly weren't running fast, especially if you run them with a built-in BASIC interpreter. It is not surprising that I quickly learned assembler and spent lots of early years shifting bits and registers around. So did almost everybody else who wanted to release a commercial application written for one of those computers. There were, more or less, no games and applications written in BASIC simply because they would run too slow and nobody would use them.

Time has changed; computers are now fast—incredibly fast! If you don't believe me, check the code examples for this book. A lot of times, I had to write loops that spin over many million iterations so that the result of changing the code would be noticed at all. The raw speed of processors has also changed the software development world. Low-level languages such as assembler and C mostly gave way to more abstract approaches—C#, C++, Delphi, F#, Java, Python, Ruby, JavaScript, Go, and so on. The choice is yours. Almost anything you write in these languages runs fast or at least fast enough.

Computers are so fast that we sometimes forget the basic rule—performance matters. Customers like programs that operate so fast that they don't have to think about it. If they have to wait 10 seconds for a form to appear after clicking on a button, they won't be very happy. They'll probably still use the software, though, provided that it works for them and doesn't crash. On the other hand, if you write a data processing application that needs 26 hours for a job that executes daily, you'll certainly lose them.

I'm not saying that you should switch to assembler. Low-level languages are fast, but coding in them is too slow for modern times, and the probability of introducing bugs is just too high. High-level languages are just fine, but you have to know how to use them. You have to know what is fast and what not and—preferably—you should take this into account when designing the code.

This book will walk you through the different approaches that will help you write better code. Writing fast code is not the same as optimizing a few lines of your program to the extreme. Most of the time, that is in fact the completely wrong approach! However, I'm getting ahead of myself. Let the book speak for itself.

Who this book is for

This book was written for all Delphi programmers out there. You will find something interesting inside, whether you are new to programming or a seasoned old soul. I'm talking about basic stuff, about strings and arrays, lists and objects, but I'm also discussing parallel programming, memory manager internals, and object linking. There is also plenty of dictionaries, pointers, algorithmic complexities, code inlining, parameter passing, and what not.

So, whoever you are, dear reader, I'm pretty sure you'll find something new in this book. Enjoy!

What this book covers

Chapter 1, About Performance, talks about performance. We'll dissect the term itself and try to find out what users actually mean when they say that a program is performing (or not performing) well. Then, we will move into the area of algorithm complexity. We'll skip all the boring mathematics and just mention the parts relevant to programming. We will also look at different ways of finding the slow (non-performant) parts of the program, from pure guesswork to measuring tools of a different sophistication, homemade and commercial.

Chapter 2, Fixing the Algorithm, examines a few practical examples where changing an algorithm can speed up a program dramatically. In the first part, we'll look at graphical user interfaces and what we can do when a simple update to TListBox takes too long. The second part of the chapter explores the idea of caching and presents a reusable caching class with very fast implementation. In the last part, we'll revisit some code from Chapter 1, About Performance, and make it faster, again, by changing an algorithm.

--">