Title :
Safe Overclocking of Tightly Coupled CGRAs and Processor Arrays using Razor
Author :
Brant, Alexander ; Abdelhadi, Ahmed ; Sim, Douglas H. H. ; Shao Lin Tang ; Yue, Michael Xi ; Lemieux, Guy G. F.
Author_Institution :
Dept. of ECE, Univ. of British Columbia, Vancouver, BC, Canada
Abstract :
Overclocking a CPU is a common practice among home-built PC enthusiasts where the CPU is operated at a higher frequency than its speed rating. This practice is unsafe because timing errors cannot be detected by modern CPUs and they can be practically undetectable by the end user. Using a timing speculation technique such as Razor, it is possible to detect timing errors in CPUs. To date, Razor has been shown to correct only unidirectional, feed-forward processor pipelines. In this paper, we safely overclock 2D arrays by extending Razor correction to cover bidirectional communication in a tightly coupled or lockstep fashion. To recover from an error, stall wavefronts are produced which propagate across the device. Multiple errors may arise in close proximity in time and space; if the corresponding stall wavefronts collide, they merge to produce a single unified wavefront, allowing recovery from multiple errors with one stall cycle. We demonstrate the correctness and viability of our approach by constructing a proof-of-concept prototype which runs on a traditional Altera FPGA. Our approach can be applied to custom computing arrays, systolic arrays, CGRAs, and also time-multiplexed FPGAs such as those produced by Tabula. As a result, these devices can be overclocked and safely tolerate dynamic, data-dependent timing errors. Alternatively, instead of overclocking, this same technique can be used to `undervolt´ the power supply and save energy.
Keywords :
field programmable gate arrays; systolic arrays; Altera FPGA; CPU; Razor correction; bidirectional communication; custom computing arrays; data-dependent timing errors; feedforward processor pipelines; overclocking; processor arrays; proof-of-concept prototype; systolic arrays; tightly coupled CGRA; time-multiplexed FPGA; timing speculation technique; unidirectional processor pipelines; Clocks; Delays; Pipelines; Random access memory; Registers; Synchronization;
Conference_Titel :
Field-Programmable Custom Computing Machines (FCCM), 2013 IEEE 21st Annual International Symposium on
Conference_Location :
Seattle, WA
Print_ISBN :
978-1-4673-6005-0
DOI :
10.1109/FCCM.2013.63