If your problem is that you're not happy with your work, what you need to do is find better work. There are innumerable ways of finding interesting work to do i

If your problem is that you're not happy with your work, what you need to do is find better work. There are innumerable ways of finding interesting work to do in computer science. A PhD is just one way of doing this but keep in mind that there are many other easier ways. If just applying for other jobs doesn't work, you can hack on open source projects, build your own lifestyle business, or get involved with startups. I dunno about academia, but I'm pretty sure the top-tier labs will hire you after a PhD if your research is good enough (i.e., enough papers in top-tier conferences) regardless of your age. For instance, Sriram Rajamani, who leads the program analysis group at MSR India, spent several years writing code before he started his PhD. I'd like to reiterate that a PhD is not for the faint of heart. It's hard work, difficult work and you really need to be motivated to do top-quality work for 60 hours a week at near-sustenance-level pay while competing with the smartest people in the world to publish your research.