Check out the new USENIX Web site. next up previous
Next: Register Allocation Up: Performance Previous: Performance comparison of different

Optimizations for the FastVM on x86

This section investigates the effects of the different code optimizations we implemented for the FastVM on x86. These optimizations are register allocation, method inlining, and floating-point precision mode optimizations as we described in section 3. All three optimization techniques impose a negligible amount of additional compilation time to the VM.