ABSTRACT
The MIPS R6000 microprocessor relies on a new type of translation lookaside buffer — called a TLB slice — which is less than one-tenth the size of a conventional TLB and as fast as one multiplexer delay, yet has a high enough hit rate to be practical. The fast translation makes it possible to use a physical cache without adding a translation stage to the processor's pipeline. The small size makes it possible to include address translation on-chip, even in a technology with a limited number of devices.
The key idea behind the TLB slice is to have both a virtual tag and a physical tag on a physically-indexed cache. Because of the virtual tag, the TLB slice needs to hold only enough physical page number bits — typically 4 to 8 — to complete the physical cache index, in contrast with a conventional TLB, which needs to hold both a virtual page number and a physical page number. The virtual page number is unnecessary because the TLB slice needs to provide only a hint for the translated physical address rather than a guarantee. The full physical page number is unnecessary because the cache hit logic is based on the virtual tag. Furthermore, if the cache is multi-level and references to the TLB slice are “shielded” by hits in a virtually indexed primary cache, the slice can get by with very few entries, once again lowering its cost and increasing its speed. With this mechanism, the simplicity of a physical cache can been combined with the speed of a virtual cache.
- Bcck84.John Beck ct. al., "A 3b Microprocessor with On-Chip Virtual Memory Management," 1984 IEEE Inrernational Solid Stnte Circlrits Conference, pp. 17- 170.Google Scholar
- Cheng87.Ray Cheng, Virtual Address Cache in UNIX, Proceedings of Summer 1987 USENIX Conference, pp. 217-224.Google Scholar
- Denning70.Peter J. Denning, Virtual Memory, Computing Surveys, vol. 2, no. 3, September 1970. Google ScholarDigital Library
- Goodman87.James R. Goodman, Coherency for Multiprocessor Virtual Address Caches, Proc. Second International Conference on Architectural Support for Programming Languages and Qerating Systems, October 1987, pp. 72-81. Google ScholarDigital Library
- Hill86.Mark Hill et. al., Design Decisions in SPUR, Computer vol. 19, no. 11, November 1986, pp. 8-22. Google ScholarDigital Library
- Kane87.Gerry Kane, MIPS RISC Architecture, Prentice-Hall, 19S7. Google ScholarDigital Library
- Lee60.F. F. Lee, Study of 'Look Aside' Memory, IEEE Transactions on Computers, vol. lS, no. 11, November 1960, pp. 1062-1064.Google Scholar
- Lee89.Ruby Lee, Precision Architecture, Computer, vol. 22, 110. 1, January 1989, pp. 78-91. Google ScholarDigital Library
- Przybylski89.Steven Przybylski, Mark Horowitz and John Hennessy, Characteristics of Performance-Optimal Multi-Level Cache Hierarchies, PrOC. Sixteenth IEEE/ACM International Sympositrm on Computer Architecture, June 1989, pp. 114-121. Google ScholarDigital Library
- Riordan89.Torn Riordan, G.P. Grewel, Simon Hsu, John Kinsel, Jeff Libby, Roger March, Marvin Mills, Paul Ries and Rancly Scoikld, System Design Using the MIPS R3OCO/3010 RISC Chipset, Digest o,f Papers Spring 1989 IEEE Compcon, pp. 494498.Google Scholar
- Roberts90.David Roberts, Tim Layman and George Taylor, An ECL Microprocessor Designed for Two-Level Cache, Digest of Papers Spring 1990 IEEE Compcon, pp. 228-231.Google Scholar
- Short88.Robert Short and Henry Levy, A Simulation Study of Two-Level Caches, Proc. Fifteenth IEEEIACM International Symposium on Computer Architecture, June 1988, pp. 81-88. Google ScholarDigital Library
- Smith82.Alan J. Smith, Cache Memories, Computing Surveys, vol. 14, no. 3, September 1982, pp. 473-530. Google ScholarDigital Library
- Wang89.Wen-Harm Wang, Jean-Loup Baer and Henry Levy, Organization and Performance of a Two-Level Virtual-l&l Cache Hierarchy, Proc. Skteenth IEEE/ACM International Simpositrum on Computer Architecture, June 19S9, pp. 140-14s. Google ScholarDigital Library
- Wood86.David Wood et. al., An In-Cache Address Translation Mechanism, Proc. Thirteenth. IEEE/ACM International Symposium on Computer Architecture, June 1986, pp. 358365. Google ScholarDigital Library
Index Terms
- The TLB slice—a low-cost high-speed address translation mechanism
Recommendations
The TLB slice—a low-cost high-speed address translation mechanism
Special Issue: Proceedings of the 17th annual international symposium on Computer ArchitectureThe MIPS R6000 microprocessor relies on a new type of translation lookaside buffer — called a TLB slice — which is less than one-tenth the size of a conventional TLB and as fast as one multiplexer delay, yet has a high enough hit rate to be practical. ...
TLB Improvements for Chip Multiprocessors: Inter-Core Cooperative Prefetchers and Shared Last-Level TLBs
Translation Lookaside Buffers (TLBs) are critical to overall system performance. Much past research has addressed uniprocessor TLBs, lowering access times and miss rates. However, as Chip MultiProcessors (CMPs) become ubiquitous, TLB design and ...
Synergistic TLBs for High Performance Address Translation in Chip Multiprocessors
MICRO '43: Proceedings of the 2010 43rd Annual IEEE/ACM International Symposium on MicroarchitectureTranslation Look-aside Buffers (TLBs) are vital hardware support for virtual memory management in high performance computer systems and have a momentous influence on overall system performance. Numerous techniques to reduce TLB miss latencies including ...
Comments