Ted Baker (a CS professor at FSU) suggested I look at a paper by Bertogna, and Cirinei from the 2007 RTSS that pushes the state of the art for RT analysis of MP systems. I can't find a fully-accessible version of the paper on-line but here's the ACM portal page for the paper. I'm a bit excited about this new analysis algorithm.

First, this paper uses an MP version of Response Time Analysis, the same approach I use for uniprocessor analysis. That means it will give the worst-case response time for each task, not just feasible/infeasible for the entire system. Second, it uses a trick that I've been seeing. It wants to look at the target system at a resolution of scheduling clock ticks. This technique may improve the efficiency of the algorithm, but it looks to me like the analysis in the paper might not work if all time measures are real numbers. This clock-tick view of scheduling struck me because I've been considering another analysis algorithm for the RT analysis calculator. It's a mechanism from a paper by Ted Baker that is based on an exhaustive search of scheduling space. It has a branch in the search tree at every scheduling tick.