Enter your mobile number or email address below and we'll send you a link to download the free Kindle App. Then you can start reading Kindle books on your smartphone, tablet, or computer - no Kindle device required.
To get the free app, enter your e-mail address or mobile phone number.
|Print List Price:||£23.50|
Save £1.18 (5%)
Android on x86: An Introduction to Optimizing for Intel Architecture Kindle Edition
|New from||Used from|
|Length: 384 pages|
Customers Who Bought This Item Also Bought
More About the Author
What Other Items Do Customers Buy After Viewing This Item?
Most Helpful Customer Reviews on Amazon.com (beta)
The first few chapters cover the History and evolution of Android, and some discussion of the chips used to power the systems, along with a market evolution and discussion. Chapter 5 includes an interesting discussion of Intel x86 vs Arm and then does a nice job of covering x86 architecture. The practical need to build an x86 version for best performance is discussed along with the mention of a binary translator which assures all ARM code will run on x86 systems (pg 46 for picture and description).
Chapter 6 Steps you through installing the Android SDK and covers emulation if you don't have destination hardware for testing your builds.
Chapter 7 gets into the main reason for the book, how to create and port an NDK application. Starting with a discussion of Java and using JNI to call C code, then stepping into NDK application development and the use of the Android NDK tools, Cygwin and steps to build your app and then NDK compiler optimizations, including coverage of the Intel specific switches (great summary to keep handy).
Chapter 8 covers debugging and using ADB. Very briefly there are mentions of the Intel GPA tool, JTAG debugging, device driver debugging, a brief discussion of the differences between ARM and x86 (CISC vs RISC) and one paragraph on hyper threading.
Chapter 9 is the largest chapter in the book and where we get to the details performance optimizations. They cover types of optimization, such as optimizing for speed, code size or power consumption. Tools are again discussed, including GPA (18 pages) and VTune (10 pages). Steps are shown for using the tools but little into understanding the data and charts. There's 40 pages on doing a thread analysis and linking that back to the tools also.
Chapter 10 looks at vectorization, IPP and steps through a sample as optimized several ways for comparison.
Chapter 11 looks at using HAXM and optimizing your virtual devices.
Chapter 12 finished the book by discussing performance testing and profiling apps along with platform turning.
The examples and data shown were captured during Medfield timeframe so there's new hardware coming out that dates this now. It is better than anything I've seen in the public space up to now.
5 stars, I highly recommend this book as a great reference for learning and doing performance tuning, for improving your apps and taking advantage of x86 features that can set your program apart from others by superior performance. As with all things programming wise, you'll have to understand and build on this but it's a great reference and start for your tool box of goodies.
This would have been a five star but for the fact that it can be a little too technical at times. Even with my I.T. background there are times when I had to reread something several times. Still, fantastic guidebook.