Branch recovery with compiler-assisted multiple instruction retry
- 2 January 2003
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
A compiler-assisted approach to implementing multiple instruction retry has recently been developed by C.-C. J Li et al. (1991). They extend compiler-assisted multiple instruction retry to include a broad class of code execution failures. Five benchmarks were used to measure the performance penalty of hazard resolution. Results indicate that the enhanced pure software approach can produce performance penalties consistent with existing hardware techniques. A combined compiler/hardware resolution strategy is also described and was evaluated. Experimental results indicate a lower performance penalty than with either a totally hardware or totally software approach.Keywords
This publication has 8 references indexed in Scilit:
- The UCLA mirror processor: a building block for self-checking self-repairing computing nodesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Checkpointing and rollback-recovery in distributed object based systemsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- High-performance fault-tolerant VLSI systems using micro rollbackIEEE Transactions on Computers, 1990
- Implementing precise interrupts in pipelined processorsIEEE Transactions on Computers, 1988
- Checkpoint Repair for High-Performance Out-of-Order Execution MachinesIEEE Transactions on Computers, 1987
- A study of scalar compilation techniques for pipelined supercomputersACM SIGARCH Computer Architecture News, 1987
- Resilient Distributed ComputingIEEE Transactions on Software Engineering, 1984
- System Development and Technology Aspects of the IBM 3081 Processor ComplexIBM Journal of Research and Development, 1982