I was studying the subject test last year. Here is the list of books I used, which were pretty much the textbooks for my math classes at my undergraduate institution:
1. Calculus by Stewart.
2. Elementary Differential Equations by Boyce and DiPrima.
3. Principles of Analysis by Rudin. (This book is a classic, but analysis is not a major area in the test)
4. Fundamentals of Complex Analysis by Saff and Snider.
5. Topology by Munkres. Introduction to Topology by Gamelin and Greene is also a decent one and considerably cheaper. I believe it is only around $15 comparing to over $100 for Munkres....
6. Linear Algebra by Freidberg, Insel, and another author whom I cannot remember.
7. Algebra: An Introduction by Hungerford. (Dummit and Foote book is more well-written. I suggest that).
However, Topology, analysis, and other topics are only 25% of the test, and questions are only at the most fundamental level. Use these books with caution as they cover much more materials and greater depth than GRE.
There are pros and cons about the REA book and Princeton Review book, which are pretty much all you can find on the market today. Problems in Princeton Review are much easier than those in actual tests, and some areas are omitted in their chapter reviews. They are also much more computational than the theoretical ones in the actual exam. That being said, it is still a good guideline about what you should expect on the test. REA book has many weird problems that will unlikely appear on the test, and some of them are pretty hard. Chapter reviews are well written in the REA book. I used PR as the major tool with REA as the supplement. My suggestion is that you do as many problems in the textbooks as you can on top of the REA and PR books.
One thing you should note is that there are some typos and wrong answers in Chapter 7 of the PR book. I believe there is a list of them here in the forum. You might want to search them.
I hope this served you well and good luck!