Mechanism for fast access to control space in a pipeline processor

(54) MECHANISM FOR FAST ACCESS TO CONTROL SPACE IN A PIPELINE PROCESSOR

(75) Inventors: Margaret Gearty, Bathford Bath (GB);

Chih-Jui Peng, San Jose, CA (US)

(73) Assignee: Hitachi, Ltd., Tokyo (JP)

( * ) Notice: Subject to any disclaimer, the term ol this patent is extended or adjusted under 35 U.S.C. 154(b) by 0 days.

(21) Appl. No.: 09/410,926

(22) Filed: Oct. 1, 1999

(51) Int. C I. G06F 9/30

(52) U.S. C I 712/225

(58) Field of Search 712/209, 225,

712/233

(56) References Cited

U.S. PATENT DOCUMENTS

4,814,981 A 3/1989 Rubinfeld 364/200

5,142,634 A * 8/1992 Fite et al 712/240

5,251,311 A 10/1993 Kasai 395/425

5,386,565 A 1/1995 Tanaka et al 395/700

5,423,050 A 6/1995 Taylor et al 395/575

5,434,804 A 7/1995 Bock et al 364/579

5,440,705 A 8/1995 Wang et al 395/421.1

5,448,576 A 9/1995 Russell 371/22.3

5,452,432 A 9/1995 Macachor 395/425

5,455,936 A 10/1995 Maemura 395/183.11

5,479,652 A 12/1995 Dreyer et al 395/183.06

5,483,518 A 1/1996 Whetsel 370/13

5,488,688 A 1/1996 Gonzales et al 395/183.1

5,530,965 A 6/1996 Kawasaki et al 395/800

5,570,375 A 10/1996 Tsai et al 371/22.3

5,590,354 A 12/1996 Klapproth et al 395/800

5,596,734 A 1/1997 Ferra 395/825

5,598,551 A 1/1997 Barajas et al 395/484

5,608,881 A 3/1997 Masumura et al 395/306

5,613,153 A 3/1997 Arimilli et al 395/821

(List continued on next page.)

A method for low latency access to the control space. A pipeline processor executes instructions in multiple stages including a decode stage, one or more execution, stages, and a writeback stage. A control space access instruction includes a first field containing a control register specifier and a second field containing a general purpose register specifier. The decode stage is configured to decode the first and second fields and place the decoded contents on a global operand bus. The specified control register is addressed from the global operand bus while the access instruction is in decode. In the case of a read instruction, the addressed control register places its contents on the global operand bus while the instruction remains in decode. In the case of a write instruction, the general purpose register is addressed during the execution stage and its contents placed on the global operand bus during the writeback stage such that the contents of the addressed general purpose register are moved to the addressed configuration register during the writeback stage.

7 Claims, 9 Drawing Sheets

5,884,092 A 3/1999 Kiuchi et al 395/800.35

5,896,550 A 4/1999 Wehunt et al 395/846

5,918,045 A 6/1999 Nishii et al 395/584

5,930,523 A 7/1999 Kawasaki et al 395/800.32

5,930,833 A 7/1999 Yoshioka et al 711/210

5,944,841 A 8/1999 Christie 714/38

5,950,012 A 9/1999 Shiell et al 395/712

5,953,538 A 9/1999 Duncan et al 395/842

5,956,477 A 9/1999 Ranson et al 395/183.06

5,978,874 A 11/1999 Singhal et al 710/107

5,978,902 A 11/1999 Mann 712/227

5,983,017 A 11/1999 Kemp et al 395/704

5,983,379 A 11/1999 Warren 714/727

6,014,734 A * 1/2000 Tran et al 712/23

FOREIGN PATENT DOCUMENTS

JP PCT/JP96/02819 9/1996 G06F/9/46

JP 8320796 A 12/1996 G06F/9/46

JP 8329687 A 12/1996 G11C/15/00

JP 9212358 A 8/1997 G06F/9/38

JP 9311786 A 12/1997 G06F/9/38

JP 10106269 A 4/1998 G06F/12/08

JP 10124484 A 5/1998 G06F/17/10

JP 10177520 A 6/1998 G06F/12/10

* cited by examiner