Removing redundant information in hybrid branch prediction

• US 20030065912A1
• Filed: 10/03/2001
• Published: 04/03/2003
• Est. Priority Date: 10/03/2001
• Status: Active Grant

• Current Assignee: Micron Technology, Inc.
• Original Assignee: Intel Corporation
• Inventors: Jourdan, Stephan J., Hum, Herbert H.J.
• Granted Patent: US 7,203,825 B2
• Time in Patent Office: Days
• US Class Current: 712/239
• CPC Class Codes: G06F 9/3848 using hybrid branch prediction, e.g. selection between prediction techniques

Abstract

A hybrid branch predictor is disclosed. The predictor includes prediction aiding information, a plurality of branch predictors to provide a plurality of branch predictions, a plurality of storage elements to hold less than full extent of the branch predictions, but sharing information among said plurality of storage elements enables extraction of said full extent of the prediction. The predictor also includes a selection mechanism to select a prediction from the plurality of branch predictions.

Claims

1. A hybrid branch predictor system, comprising:

prediction aiding information;

a plurality of branch predictors to provide a plurality of branch predictions, each predictor to make a corresponding prediction based on said prediction aiding information;

a plurality of storage elements, each storage element of said plurality of storage elements to hold less than full extent of said corresponding prediction, but sharing information among said plurality of storage elements enables extraction of said full extent of said corresponding prediction; and

a selection mechanism to select a prediction from said plurality of branch predictions.

2. The system of claim 1, wherein said corresponding prediction includes at least one target address.

3. The system of claim 2, wherein said corresponding prediction also includes a “

taken”

/“

not take”

bit.

4. The system of claim 1, wherein said full extent of said corresponding prediction includes 32-bit address.

5. The system of claim 4, wherein said each storage element is configured to be less than or equal to 32 bits.

6. The system of claim 1, wherein said prediction aiding information includes prior outcome information.

7. The system of claim 1, wherein said prediction aiding information includes a type of branch.

8. The system of claim 1, wherein said prediction aiding information includes a correlation factor.

9. The system of claim 1, wherein said prediction aiding information includes a confidence level.

10. The system of claim 1, wherein said selection mechanism includes a counter to determine which predictor is currently more accurate.

11. The system of claim 1, wherein a predetermined number of most significant bits in said corresponding prediction is shared.

12. A method, comprising:

configuring less storage space for predictions in the hybrid branch prediction than required memory space for a single-scheme predictor; and

leveraging said predictions such that information for a selected prediction includes substantially similar amount of information as that for the single-scheme predictor.

13. The method of claim 12, wherein said leveraging includes sharing said predictions.

14. The method of claim 13, wherein said sharing predictions includes using less than full extent of one of said predictions, and borrowing from another of said predictions to produce the full extent of said information for said selected prediction.

15. A computer readable medium containing executable instructions which, when executed in a processing system, causes the system to leverage information in a hybrid branch prediction, comprising:

configuring less storage space for predictions in the hybrid branch prediction than required memory space for a single-scheme predictor; and

leveraging said predictions such that information for a selected prediction includes substantially similar amount of information as that for the single-scheme predictor.

16. The medium of claim 15, wherein said leveraging includes sharing said predictions.

17. The medium of claim 16, wherein said sharing predictions includes using less than full extent of one of said predictions, and borrowing from another of said predictions to produce the full extent of said information for said selected prediction.