Optimizing hardware TLB reload performance in a highly-threaded processor ...

(73) Assignee: Sun Microsystems, Inc., Santa Clara, CA(US)

( * ) Notice: Subject to any disclaimer, the term of this patent is extended or adjusted under 35 U.S.C. 154(b) by 164 days.

(21) Appl.No.: 10/880,985

(22) Filed: Jun. 30, 2004

(51) Int. CI.

G06F12/10 (2006.01)

(52) U.S. CI 711/204; 711/150; 711/151;

711/158; 711/202; 711/203; 711/205; 711/206;

711/216

(58) Field of Classification Search None

See application file for complete search history.

(56) References Cited

U.S. PATENT DOCUMENTS

5,446,854 A 8/1995 Khalidi et al 395/401

5,465,337 A 11/1995 Kong 395/417

5,479,627 A 12/1995 Khalidi et al 395/415

5,526,504 A * 6/1996 Hsuetal 711/207

5.712.791 A 1/1998 Lauterbach 364/489

5,717,885 A * 2/1998 Kumar et al 711/207

5.742.792 A 4/1998 Yanai et al.
5,761,707 A 6/1998 Aiken etal.

5,802,341 A * 9/1998 Kline et al 711/209

5,802,561 A 9/1998 Favaetal.

A method and apparatus for improved performance for reloading translation look-aside buffers in multithreading, multi-core processors. TSB prediction is accomplished by hashing a plurality of data parameters and generating an index that is provided as an input to a predictor array to predict the TSB page size. In one embodiment of the invention, the predictor array comprises two-bit saturating up-down counters that are used to enhance the accuracy of the TSB prediction. The saturating up-down counters are configured to avoid making rapid changes in the TSB prediction upon detection of an error. Multiple misses occur before the prediction output is changed. The page size specified by the predictor index is searched first. Using the technique described herein, errors are minimized because the counter leads to the correct result at least half the time.

20 Claims, 9 Drawing Sheets

U.S. PATENT DOCUMENTS

6,016,542 A 1/2000 Gottlieb et al 712/225

6,044,446 A 3/2000 Joy et al 711/203

6,092,154 A 7/2000 Curtis etal 711/137

6,098,169 A 8/2000 Ranganathan 712/227

6,138,209 A * 10/2000 Krolaketal 711/128

6,141,692 A 10/2000 Loewenstein et al 709 234

6,247,121 Bl 6/2001 Akkaryetal 712/239

6,272,520 Bl 8/2001 Sharangpani et al 709/108

6,295,600 Bl 9/2001 Parady 712/228

6,308,279 Bl 10/2001 Toll et al 713/323

6,314,563 Bl 11/2001 Agesenetal 717/9

6,374,349 Bl * 4/2002 McFarling 712/239

6,408,368 Bl 6/2002 Parady 711/159

6,535,905 Bl 3/2003 Kalafatis et al 709/108

6,574,725 Bl* 6/2003 Kranich et al 712/31

6,578,137 B2 6/2003 Parady 712/228

6,581,140 Bl* 6/2003 Sullivan et al 711/128

6,609,193 Bl 8/2003 Douglas et al 712/219

6,625,654 Bl 9/2003 Wolrichetal 709/230

6,633,984 B2 10/2003 Susseretal 713/201

6,647,458 Bl 11/2003 Glynn

6,671,707 Bl 12/2003 Hudson etal 707/206

6,700,410 B2 3/2004 Ebergen 326/93

6,700,825 Bl 3/2004 Ebergen 365/221

2001/0047468 Al 11/2001 Parady 712/228

2002/0052926 Al 5/2002 Bush etal 709/217

2002/0078122 Al * 6/2002 Joy etal 709/102

2002/0129309 Al 9/2002 Floyd etal 714/724

2002/0199070 Al 12/2002 Chaudhry et al.

2003/0088610 Al 5/2003 Kohnetal 709/107

2003/0088713 Al 5/2003 Mandaletal.

2003/0097518 Al 5/2003 Kohnetal 711/5

2003/0105756 Al 6/2003 Daynes 707/8

2003/0105907 Al 6/2003 Kohnetal 710/305

2003/0120825 Al 6/2003 Awari et al 709/316

2003/0149861 Al 8/2003 Becker 712/214

2003/0163645 Al 8/2003 Tremblayetal 711/137

2003/0191927 Al 10/2003 Joy etal 712/228

2003/0212874 Al 11/2003 Alderson 711/170

2004/0002974 Al 1/2004 Kravitz et al 707/8

2004/0003208 Al 1/2004 Damron 712/225

2004/0003211 Al 1/2004 Damron 712/228

2004/0006633 Al 1/2004 Chandra etal 709/231

2004/0034853 Al 2/2004 Gibbons etal 717/174

2004/0073906 Al 4/2004 Chamdani et al 718/102

2004/0117600 Al 6/2004 Bodas et al. 2004/0123028 Al 6/2004 Kanai et al. 2005/0027961 Al * 2/2005 Zhang 711/207

OTHER PUBLICATIONS

R. Heald et al., "A Third-Generation SPARC V9 64-b Microprocessor," IEEE JSSC, Nov. 2000, pp. 1526-1538.

"Igniting a Computing Revolution! Throughput Computing," Sun Microsystems Brochure.

"Introduction to Throughput Computing," Sun Microsystems whitepaper, Feb. 2003.

D. Wendell et al., "A 4 MB On-Chip L2 Cache for a 90nm 1.6GHz 64 bit Microprocessor," IEEE JSSC (Feb. 2004). G. Konstadinidis et al., "Implementation of a Third-Generation 1.1GHz 64b Microprocessor," 2002 IEEE Journal of Solid-State Circuits, vol. 37, No. 11, pp. 1461-1469 (Nov. 2002). K. Krewell, "Sun Weaves Multithreaded Future," Microprocessor Report, Reed Electronics Group (Apr. 14, 2003). "UltraSPARC® IV Processor," User's Manual Supplement, Sun Microsystems, version 1.0 (Apr. 2004). "UltraSPARC® IV Processor," Datasheet, Sun Microsystems. "The UltraSPARC® Illi Processor" Architecture Overview, Technical Whitepaper, Sun Microsystems, version 1.2 (Apr. 2004). "UltraSPARC® IV Processor," Architecture Overview, Technical Whitepaper, Sun Microsystems, Version 1.0 (Feb. 2004). R. Ronen, "VLSI Architecture Design Course," Lecture #10, Multithreaded Architectures (dated May 2003). K. Mai et al., "Smart Memories: A Modular Reconfigurable Architecture," Stanford University Proceedings of the 27th ISCA (Jun. 2000).

L. Hammond et al., "The Stanford Hydra CMP," IEEE Micro, pp. 71-84, (Mar.-Apr. 2000).

L. Hammond et al., "Considerations in the Design of Hydra: A Multiprocessor-on-a-Chip Microarchitecture," Technical Report No. CSL-TR-98-749, pp. 1-10 (Feb. 1998).

K. Privitt, et al., "Threading: Connecting the Pieces, System Architecture, Application, OS and Tools," Intel Developer Forum (Feb. 17-19, 2004).

Watanabe, K. et al. (2001) "Exploiting Java Instruction/Thread Level Parallelism with Horizontal Multithreading," IEEE, pp. 122-129. Wikipedia, "Cache," Version as of Dec. 30, 2007. Accessed Jan. 4, 2008, http://en.wikipedia.org/wiki/Cache.

Koopman, Philip, "Cache Organization," lecture material, Carnegie Mellon University, Sep. 2, 1998, http://www.ece.cmu.edu/7ece548/ handouts/04cachor.pdf.

* cited by examiner