https://github.com/dendibakh/perf-book

The book "Performance Analysis and Tuning on Modern CPU"
https://github.com/dendibakh/perf-book

Last synced: 22 days ago
JSON representation

The book "Performance Analysis and Tuning on Modern CPU"

• Host: GitHub
• URL: https://github.com/dendibakh/perf-book
• Owner: dendibakh
• License: cc0-1.0
• Created: 2021-12-06T18:45:43.000Z (almost 3 years ago)
• Default Branch: main
• Last Pushed: 2024-09-26T17:27:19.000Z (about 1 month ago)
• Last Synced: 2024-10-02T00:24:09.299Z (about 1 month ago)
• Language: TeX
• Homepage: https://book.easyperf.net/perf_book
• Size: 67.3 MB
• Stars: 2,120
• Watchers: 62
• Forks: 156
• Open Issues: 11
• Metadata Files:
  • Readme: README.md
  • Funding: .github/FUNDING.yml
  • License: LICENSE

Awesome Lists containing this project

• awesome-gemm - perf-book

README

        
[![X (formerly Twitter) Follow](https://img.shields.io/twitter/follow/dendibakh)](https://twitter.com/dendibakh)

![GitHub Repo stars](https://img.shields.io/github/stars/dendibakh/perf-book)

# perf-book

This is a repository with source files of the book "Performance Analysis and Tuning on Modern CPU" by Denis Bakhvalov, et al.

**Second edition work in progress!** Planned changes are outlined in the google [document](https://docs.google.com/document/d/1tr2qRDe72VSBYypIANYjJLM_zCdPB6S9m4LmXsQb0vQ/edit?usp=sharing). The planned new table of contents is in [new_toc.md](new_toc.md).

My goal is to accumulate as much knowledge as possible from all the best experts in the industry. And of course, share that knowledge with you. Contributions are welcome.

# Contributing

There are many ways how you can help.

- You can author a section(s) on a topic on which you are expert; let me know before you start.

- Small improvements are welcome without prior approval; just open a new PR.

- Feel free to propose ideas for new content.

- Reviewers with all backgrounds are needed.

Check out the [discussions](https://github.com/dendibakh/perf-book/discussions) page to start.

For examples on how to add images, table, code listings, etc., see [how-to.md](how-to.md).

# Building a book (pdf)

At the moment, building a PDF only works on Windows and Linux. MacOS requires building some components (e.g. pandoc-crossref from sources).

Requirements:

 * Python3. Install natsort module: `pip install natsort`.

 * [Pandoc](https://pandoc.org/installing.html). Install [version 2.9](https://github.com/jgm/pandoc/releases/tag/2.9.2.1)

 * `pandoc-fignos` and `pandoc-tablenos`. Run `pip install pandoc-fignos pandoc-tablenos`.

 * `pandoc-crossref`. This one requires manual installation. I just downloaded the binary from [here](https://github.com/lierdakil/pandoc-crossref/releases/tag/v0.3.6.4) and copied it to the same place where `pandoc-fignos` is.

 * [MiKTeX](https://miktex.org/download). Check `Yes` for automatic package installation.

Run:

```bash

# Linux bash and Windows cmd prompt

python export_book.py && pdflatex book.tex && bibtex book && pdflatex book.tex && pdflatex book.tex

# Windows powershell

function Run-Block-With-Error($block) {

    $ErrorActionPreference="Stop"

    Invoke-Command -ScriptBlock $block

}

Run-Block-With-Error {python.exe export_book.py; pdflatex book.tex; bibtex book; pdflatex book.tex; pdflatex book.tex}

```

As a result, `book.pdf` will be generated. The first compilation may be slow.

## License

[Creative Commons Zero v1.0 Universal](LICENSE)