Customer Reviews

Sparc Architecture, Assembly Language Programming, and C

5 star:		(4)
4 star:		(2)
3 star:		(1)
2 star:		(3)
1 star:		(3)

 

 

at best, this book is a quick gloss over a small, safe subset of sparc programming. a very pricey quick gloss. at worst, this book is an exercise in futility and frustration. the author's reliance and execessive use on the m4 macro processor is enough to make one walk into oncoming traffic. none of the examples in the book are decipherable without running through the author's library of m4 macro routines - rather, the reader is presented with an indirect representation of sparc assembly that makes concepts hard to learn. the author's misdirected aims of symbolic abstractions are ok in the context of a higher-level programming language, but are absolutely worthless in the context of assembly-level programming. through this book, one is encouraged to program sparc assembly in a high-level manner similar to C - actual pragmatic and real-world assembly programming idioms are nowhere to be found. needless to say, any reader will be sorrowfully disappointed to find that m4 is about as common as leprosy in production environments. i would be beaten like a red-headed step child if i were to incorporate any of the author's practices at work. do yourself a favor and pick up the documentation at sparc.com and leave richard p. paul to nance around with the m4 processor by himself in his more aptly title book "M4, C, and Sparc Architecture"

For anyone interested in learning about the SPARC Architecture/Instruction Set (and RISC machines in general) this book is invaluable. It is especially useful for optimizing iterative and decision making C/C++ constructs. In fact, if one follows the tenets espoused in this book, one can learn to hand optimize time-critical sections of C/C++ code that is better than that produced by gcc or cc - the aversion of the UNIX community to write any code in assembly language notwithstanding.

The book really delivers what its rather verbose title implies. That is, a really outstanding feature of the text is the way in which the author translates the standard C/C++ constructs to their low-level counterparts. He does this in stages - creating a variety of examples that progress from functional but grossly inefficient code fragments up to superbly succinct variants. I have used this book in a one semester undergraduate course at the University of Delaware for three years and have also used excerpts from it when I have taught the MIPS Architecture. There is no other book that treats RISC (or CISC) architecture from Professor Paul's relational premise, with which I totally agree. Having taught INTEL stuff for 10+ years, I firmly believe that much of its content could be effectively utilized in CISC courses. The book is also used as the secondary text in the graduate compiler course at U.D.

The book is not without flaws, most of which are because of an incredible number of typographical errors - I have counted over 60 just involving commas! Hopefully the new edition which I believe is due to be published soon will have been edited/typeset with more care. Also, there are some minor changes to the gnu software (gdb and gcc) that need to be upgraded.

I picked up this book to familiarize myself with the SPARC architecture for an upcoming project and I was extremely disappointed in the presentation of the material, both grammatically and intuitively.

First, when learning assembly language, the last thing a reader or student needs is the code to be obfuscated by a preprocessing tool such as m4. Hiding address offsets and variable alignments in nearly impossible to decipher macros is NOT helpful. This does not make it easier to learn assembly. I found myself learning more about a tool that I'll never use after finishing this book than about SPARC assembly.

Second, whoever edited the manuscript for this book should be fired. I found myself editing the book as I read so I could understand what the author was trying to say. I also found the language to be a bit obtuse in a few, unfortunately important, places.

Third, the diagrams in the book need some serious help as well. They were almost useless. Many of them made the topic being discussed more confusing. I found myself using Wikipedia or the Sparc V8 manual more than once.

All that said, the book does try to cover the important aspects of the SPARC architecture. I did get the needed information from the book, but it could have been organized and presented much better.

The book could be a great SPARC reference and tutorial book if these problems were addressed in a future edition.

For starters, the first apparent detail of this book is the glaring grammatical errors. Ok, no big deal, but still, it's an eyesore. The book is complex for the sake of complexity. Each chapter could easily be 2/3 or less the length it is now. The M4 macro is over-used and under-explained. If it used this much, a whole chapter should be devoted to it (at least more than a four page section covering few basics). The examples are poor and many of them simply don't work. A total lack of explanation as to what is actually going on "behind the scenes" as the macro does its work left me hung out to dry on many occasions. With way too much work, I reaped very little knowledge from this book. I can see this being a half-decent reference for those who have extensive knowledge of the M4 macro and previous experience in assembly language. If you're a beginner, stay miles away from this book. Books on a topic as inherently confusing as this need to be clearer and more extensive in their explanations and have examples that work.

I found the book quite useful as an introduction to SPARC assembly language. There is a lot of material on program optimization, which could be skipped or read. The extensive use of the m4 preprocessor gets slightly annoying at times, but the programs generally remain easy to read. The reason for a four star review is that although I feel the book does not cover the subject in a lot of depth, it wasn't very user-friendly to the absolute beginner either. As such, the book would make a good textbook for a course, but it is not that good for the independent learner. By the way, I'm basing this on the first edition.

This book is the text for a CS course I am taking. Its not easy to read, but neither is assembly language. No doubt, the book makes you dig in and I hate the HP assembly but its a good way to introduce stack machines. I also liked the constant comparisons to C. Bottom line - don't buy unless your ready to work your butt off.

I needed this as backup for an assembly programming course and it was a great help. It's pretty straight forward and basic, but it could go more in-depth and may not serve all that well as your only source for assembly language programming.

This book is excellent considering the time it was produced and definitely an superb piece of work.

But an update is needed as technologies have progressed much over the last 8-9 years; e.g. Sun Microsystems have incorporated the GCC compiler into their latest architectures like the x86-Opteron; to compare and offer new perspectives on the subject maybe advances in compiler technology, if any etc would be terrific.

I am sure previous reviewer have some kind of relations with the author. If this book was $15, I would say buy & keep it for reference. But, this is one of the worst books I have ever read and/or studied. If you have no idea about SPARC, it is almost impossible for you to understand or learn anything from this book. If you know fairly, you don't need this book, you can study it by examples by yourself. Do not waste your money.

I'm sorry, this book is an abomination! It makes a simple thing difficult. Students, prepare to suffer. This is one of the only SPARC books out there.