Mac OS X and iOS Internals: To the Apple's Core 1st Edition

It is true that reading a kernel book can be difficult and not fun, but the problem of this book is beyond this.If you read the table of content, it looks like this book covers a lot of aspects. Providing "Mac OS X Internal" is somewhat "outdated", this book might be the only one you can get in the market. It does cover a lot of aspects, but leaves none of them discussed in-depth. For example, VFS. It is sad that Mac OS X Internal doesn't really cover a lot on VFS. I bought this book, hoping that it could discuss more about Mac OS X VFS, but I was disappointed. It covers even less. The VFS Mac OS X is using now was a fork of FreeBSD. They share a common ancestor at some point of time, then they diverged.If you look into the current FreeBSD VFS, it's very different with the one Mac OS X is using. So you cannot simply depend on a FreeBSD kernel book. I don't want to make my code dirty, I always want to make sure I have a good understanding of all the assumptions of VFS interfaces before I move on. The fact is that, there's no single book on shelf can reach this goal. Apple doesn't have a good documentation of their VFS, either. So as an independent developer, reading Apple's existing code (devfs for memory based VFS, HFS for block device based VFS) seems to be the only way to learn it.Beside the above problem, another big issue is the typos and errors in this book.There are quite a number of typos. Most of them don't cause real trouble since you can correct them by yourselves, but some of them are really misleading. Like on page 348, the first parameter of mach_msg call should be a pointer type, but the star is missing. I stared at it for a minute and looked at the real source code, proved that the book was wrong. Other errors also make glitches while reading. Like, when talking about the internal Mach IPC mechanism, the book states that we've talked about semaphores. I was surprised, because I'm always a careful reader, but I didn't remember I've read any kernel synchronization stuff from the book. Later on I realized that the synchronization part was moved to after the internal IPC part. So the actual ordering of topics of that chapter is different from the one printed on the first page of that chapter. This kind of mistakes should have been avoided by carefully reading before publishing, but sadly that was not the true story.As a conclusion, if you are look for a book that can give you directions of possible solutions of your current questions/problems, it is okay. It also lists references at the end of each chapter. But if you are looking for some in-depth discussion, you might find some luck with Mac OS X Internal, though that book is really outdated. However, not every single problem has a perfect solution. For example, I really wanted a book that talks more about Mac OS X VFS, but there's just no such thing in the market.

This book contains substantial information about Apple OS's but also has some problems that diminish its value.First, it has too many typographical, spelling, and grammatical errors. This is common to Wrox books and I have to assume that the publisher doesn't bother with using a copy editor.Second, it suffers from factual errors. This is also common with Wrox books. Technical books like this would be served by a technical editorial review.Finally, it often suffers from a lack of explanatory detail. The blame for this is largely Apple's with their secretive ways. I find long passages that are either line-by-line descriptions of source code or (worse) multi-page copies of the code itself. If I wanted that level of detail I could read the source myself. (The code in question is open source--something I learned to my amazement from reading this book.). What's missing is analysis of why the code is doing what it's doing. What's also missing of course is any description of the vast majority of the code that is closed source. For those portions of the system the author relies on hacking and reverse engineering.I'm not saying the book is worthless. I did learn from it. But I found it endlessly frustrating as well. Blame Apple for the secrecy of the technology, but the editorial sloppiness has no excuse.UPDATE: I've received multiple comments on this review on my personal blog. The comments center around (1) 2 stars in unfairly low and (2) I should give more concrete examples of the errors I criticize. I can see the merit of both arguments. As for point (1) I have changed the number of stars from 2 to 3. As for (2) I would have to go back and re-read significant portions of the book in order to locate such examples, which I just don't have the time to do right now.

So I've got an entire bookshelf dedicated to lower level software books and this is by far the most versatile and most information packed. It is incredibly terse and comprehensive. I can see anyone from just an average user that wants to learn about macs to a senior developer who has much experience in c and even objective c but not necessarily the inner workings of Darwin find a lot of use from this book. I have not finished it yet (its almost 1000 pages) but I have already learned enough to fill 10 buckets. What I also love is how it refers to other books that go far more in depth on each topic. So yes in a sense this book is comprised of summaries but they still go in depth enough for one to understand the gist of whats going on. I can say from the start that the chapter on EFI is pretty phenomenal and that coupled with all the information on the mach architecture has taught me an incredible amount. Overall if you are interested in learning about macs and how they work, or are a programmer and want to learn how to take advantage of some lower level stuff or if you're more in the hacker field and enjoy seeing how things work under the hood and how you can manipulate them to your advantage (me) you will definitely enjoy this book.

Yes that is what you need to know about low level development on a C/C++ level of coding.Thats for me ... but maybe not for others. Apple want to hide this all under a swift layer. But this only works limited and if you never want to get cross platform.

There was only one book which gave insight  Mac OS X Internals: A Systems Approach , believe me its awesome, the only drawback, it was written at the time when Mac OS X was migrating from PPC to Intel. Fundamentally it was good but things have changed. After a very long gap Jonathan Levin came up with this marvel.Any one who is interested to know about Mac OS X under the hood should get a copy.Warning ---- this book in not for someone who started programming yesterday, neither its administrative guide nor for Application programmer. If you have an ITCH to KNOW the BUY IT, and please buy BOTH of them. Mac OS X Internals: A Systems Approach

As the author states in the Introduction: “This book aims to do for XNU what Bovet & Cesati's Understanding the Linux Kernel does for Linux, and Russinovich's Windows Internals does for Windows.” Knowing only the latter book, I would say that authors intention have been perfectly fulfilled. For those, which also knows Russinovich's and Solomon's standard work enough said.For those others: The author understands his matter. He not only describes, but also make understand. To make a picture, the potentially reader should go to [...]. There he can read the Contents of the book and also download a full text of ( very important ) chapter 4. There are also some little tools ( with source code ) for delving deeper into OS X and iOS. Giving the tempo with which operating systems evolve, the book is pretty actual ( Lion ). You have not to read it from page first to page last, but you will be often happy to have it in the bookshelf. And you have not to be an expert in C or Assembler to profit from it ( although having some elementary knowledge of C would enlarge the pleasure ).

Required reading if you do any OSX kernel development.

solved the purpose.