Dynamic instruction reuse
Avinash Sodani, Gurindar S. Sohi · Proceedings - International Symposium on Computer Architecture/Proceedings · 1997
This paper introduces the concept of dynamic instruction reuse. Empirical observations suggest that many instructions, and groups of instructions, having the same inputs, are executed dynamically. Such instructions do not have to be executed repeatedly --- their results can be obtained from a buffer where they were saved previously. This paper presents three hardware schemes for exploiting the phenomenon of dynamic instruction reuse, and evaluates their effectiveness using execution-driven simulation. We find that in some cases over 50% of the instructions can be reused. The speedups so obtained, though less striking than the percentage of instructions reused, are still quite significant.
Cited by 358 works. doi:10.1145/264107.264200 Open-access PDF
Authors
References
- Structure and Interpretation of Computer Programs
- Evaluating future microprocessors : The SimpleScalar tool set
- Instruction issue logic for high-performance, interruptible, multiple functional unit, pipelined computers
- Value locality and load value prediction
- A general approach for run-time specialization and its application to C
- Exceeding the Dataflow Limit via Value Prediction
- Implementing precise interrupts in pipelined processors
- Decoupled access/execute computer architectures
- The predictability of data values
- Highly accurate data value prediction using hybrid predictors