 |
|
Picking up Real-Time Programming from the shelf, it trumpets itself "The Complete Reference for Building 32-bit Real-Time Embedded Applications." Reading Real-time Programming, by Rick Grehan, Robert Moote, and Ingo Cyliax, one feels nostalgia, a little regret, and a smidgen of impatience with the authors.
Real-Time Programming isn't a book about real-time programming. It's a book about embedded programming in the PC environment, notwithstanding a few superficial remarks about real-time systems tossed off in the Introduction.
Real-time programming is a branch of computer control engineering. Real-time programming is a lot of math, it's a very detailed understanding of hardware, and it's you-bet-the-farm game theory. Publishers and authors claiming single-volume "complete references" to real-time programming are debasing the professional coinage.
Grehan et al's offering is, however, a tolerably rousing intro to doing neat stuff with Phar Lap's 32-bit ETS Kernel included on the accompanying CD-ROM. Some of the examples are very well chosen. I was enjoying reading Chapter 6 ("Keyboard and Screen I/O") when carelessness and a lack of service to the reader obtruded. Figure 6-5 purports to show a Hitachi LM032 LCD attached to a "CPU or Parallel I/O Chip." Why bother to tell us that it's an LM032 if you're just going to draw two bare rectangles and 12 black lines connecting them without labeling the signals? Sparseness of incidental detail is not a virtue in an embedded control tome.
Chapter 9 ("How Interrupts Work") weeps and pleads for genuine diagrams instead of simplistic scrawls. Figure 9-3 shows an interrupt arriving and suffers in comparison to diagrams 14-2 through 14-5 of Lance Leventhal et al's 1986 68000 Assembly Language Programming (McGraw-Hill ISBN 0-07-881232-1). Haven't we gained anything in 14 years of increasingly more powerful desktop publishing software?
The "Smart UPS" project is a very good narration and a clever project. It leads to a passable introduction to classic PC UART control programming, despite the careless sparse diagrams like Figure 13-6, which fails to expand the reader's horizons by labeling the bits, the signals, the bit order of transmission, the number of bits on the data bus, and the like.
The Table of Contents in itself is a good list of topics novices could busy themselves with for years until that moment of achieving single-board satori and attaining the Zen of Real Time. The narrative leaps and bounds from ROMs to networks to the HTTP protocol and back through the ISR. It's the breadth that is breath-taking and makes for the disappointment of realizing that no topic is going to receive any significant discussion except the tools. Well, that's the life raft when you're cast adrift--make sure you know how the tools work.
What should Grehan et al's opus have been called? I was tempted to suggest Big Bedtime Book for Young Embedded Programmers but it doesn't have enough pretty pictures, so I settled on Phar Lap Introductory Guide to Embedded Development. That would have been an excellent title to describe this medium-well-done embedded control burger.
-- Jack Woehr
| Readability |
|
| Originality |
|
| Organization |
|
| Accuracy |
|
| Consistency |
|
| Depth |
|
| Timeliness |
|
| Editing |
|
| Design |
|
| Overall Value |
|
Explanation of ERCB rating scale: No stars = unacceptable, 1 Star = marginal, 2 Stars = average, 3 Stars = above average, 4 Stars = exceptional.