Thomas_M.__Conte
Accurate and practical profile-driven compilation using the profile buffer
Dynamic rescheduling: A technique for object code compatibility in VLIW architectures
Optimization of instruction fetch mechanisms for high issue rates
Reducing State Loss for Effective Trace Sampling of Superscalar Processors
Systematic Prototyping of Superscalar Computer Architectures
Tradeoffs in processor/memory interfaces for superscalar processors
Using Branch Handling Hardware to Support Profile-Driven Optimization