How useful/applicable is the MINIX book?

How useful/applicable is the MINIX book?

I feel like I have a pretty good grasp of C and just finished the first chapter, which was useful because MINIX conforms to the POSIX standard, but before I go and read the rest I want to know if the info past here is applicable to most UNIX OSs or if it is very specific to MINIX.

My goal is to understand OS design but I don't want to go ahead if all the info in the book pertains only to MINIX. Is this a good place to start learning about OSs or is there another book/source for better information?

Other urls found in this thread:

harmful.cat-v.org/software/andy_tanenbaum
en.wikipedia.org/wiki/Genera_(operating_system)
twitter.com/NSFWRedditVideo

relevant to my interests too

i heard good things about this book op, but i havent' started it

this is a good book, worth reading if you're interested in operating systems, even modern operating systems. it's so basic you probably shouldn't worry about the content not being relevant.

thanks I'll look into this! There's a book I have but haven't started about the Linux kernel specifically, but it kinda assumes some prior knowledge which is why I moved on to the MINIX book for the time being.

Tanenbaum has great books, read computer networks few months ago.

I miss when unix book covers were whimsical and had nothing to do with the contents.

That's not a UNIX book.

Why not read superior Abraham Silberschatz's book?

I didn't say it was.

what is better about this book? just curious before I get to into MINIX

If you want theory then get "Operating System Concepts" by Silberschatz, Galvin, and Gagne or "Modern Operating Systems" by Tanenbaum
If you want something dated on the design philosophy of UNIX then get "The Design of the UNIX Operating System" by Bach
If you want something modern then get "The Design and Implementation of the FreeBSD Operating System" by McKusick, Neville-Neil, and Watson

I haven't actually read the Minix book, but I do OS development IRL and would say with certainty that silberschatz's book will do almost nothing to teach you how to actually write an OS. Silbershatz teaches you what an OS does and why it's important, but won't help you actually put one together. I don't know if the Minix book goes more in-depth with code or whatever, but if it does, I'd highly recommend something like that over silberschatz.

There's no better way to learn OS stuff than by actually looking at OS code. Abstract shit like Silberschatz has very little place in most OS work.

theory is for the architects who instruct the engineers. practice is for engineers who come up with the code

Jesus shit, it's on the ninth edition? Man, they must be really after some shekels. There really haven't been any developments in OS in the past 5 years that belong in an undergrad textbook. OP, if you get this book, do not by any means pay extra for a newer version. If anything, don't pay for it at all.

so from a security perspective then MINIX is better because it actually goes in depth about how the code works to bring the OS together???

I did a paper at university where that was the textbook.
I didn't read the whole thing, but the parts I did read were pretty good.

Maybe if you work in design-by-committee management hell. A good programmer will know both theory and be able to program well. Someone who has no idea about actual code sure as hell wouldn't be able to give reasonable OS design guidelines to someone familiar with hardware and OS programming. For example, see the Pfair scheduler (for real-time, mostly). Beautiful in theory, impossibly high overhead in practice.

Also, I'd bet that very few developers for the core parts of the Linux kernel would like it very much if some theory guy who didn't know how to program told them what to do. Of course, if that theory guy learns how to program and can maintain his own code, that's great (and is likely the case with most open-source OS devs).

If minix goes over the code, then absolutely. Security without code is management-speak. Don't even teach yourself to think like that unless you plan to work for IBM or the government.

Andrew wrote the book so it would be applicable to OSes in general, Linus Torvalds actually learned about OS design from studying MINIX

This should be good. The Basic Kernel
(Operating System Source Code Secrets Vol. I) would also be a good read.

Oh my bad I thought you two are computer science students.
Go ahead read MINIX for step by step OS development.
Silberschatz is abstract book. Not meant for SJW tier coders who can't understand after certain level of abstraction.They need to know objects and api to use.

>Jesus shit, it's on the ninth edition

They release a new edition every ~4 years

Hardcover: 548 pages 1st edition (1983)
Hardcover: 625 pages 2nd edition (1985)
Hardcover: 696 pages 3rd edition (October 1990)
Hardcover: 780 pages 4th edition (January 1994)
Hardcover: 912 pages 5th edition (January 1, 1998)
Hardcover: 912 pages 6th edition (June 26, 2001)
Hardcover: 921 pages 7th edition (December 14, 2004)
Hardcover: 992 pages 8th edition (July 29, 2008)
Hardcover: 976 pages 9th edition (December 17, 2012)

I am a computer science student but so far I'm stuck doing pre reqs full of content I've already learned

the racoon is reading the book that the raccoon is reading that the raccoon is reading that the raccoon is reading that the raccoon is reading that the raccoon is reading that the raccoon is reading that the raccoon is reading that the raccoon is reading that the raccoon is reading that the raccoon is reading

I was into osdev for a year a while back. Wish I learned about this book right from the start. It taught all the technical details on how to make an OS on i386. Linus had it, and I have no doubt this book is what allowed him to write Linux.

Reminder that real OS devs make fun of Tanenbaum.

harmful.cat-v.org/software/andy_tanenbaum

the best book is the one you read

>I am a computer science student but so far I'm stuck doing pre reqs full of content I've already learned
yeah, i did a short course at a competing uni, and am hoping to get most of my first year stuff waived for my real degree

#include
using namespace std;
int main(){
while(1)
cout

another good one is Peter Norton's Assembly Language Book for the IBM PC

Dude, I'm really angry with this to be honest. His view on OSs is actually a very good one and a logical one. But just because he once argued with Linus and Linux became popular everyone thinks they know better than him.

In reality most of the issues with Linux, biggest one I can point out is you are literally asking companies to maintain a piece of your kernel for drivers, are actually coming from points he made during the first flame war. In the end, with loadable kernel modules and stuff they came to what he said was right a million years ago. He just became the dude everyone hates because they think Linux is cool.

Man i've taken OS and i don't think you can learn any OS programming by theory, you have to code or else it's all a bunch of tangled nonsense.

that book is all mathematical logic. it's basically a mathematical proof of an OS. very hard

He didn't say you said it was.

Nigga, I'm a grad student in CS theory (as applied to scheduling, locks, etc) and I'll say for a damn fact that silberschatz is shit for learning what you actually need to know. Anybody can understand what a scheduler does, but people who are good need to be able to implement one in a real OS. A superficial, high level understanding will not get you there.

Meant for

Also, Silberschatz is the book for code artisans who only think in terms of APIs. It's for undergrads who need to check the "operating systems" box. It should be completely telling that there's a fucking *Java* edition of that book.

How do you feel knowing that we missed out on the greatest OS of all time? en.wikipedia.org/wiki/Genera_(operating_system)

ou