Electrical computers and digital processing systems: processing – Processing control – Branching
Patent
1999-01-19
2000-04-25
Kim, Kenneth S.
Electrical computers and digital processing systems: processing
Processing control
Branching
712240, 712215, G06F 938
Patent
active
060556293
ABSTRACT:
A method for a prediction correlation between a first group of branch instructions in a bunch of instructions and a second group of branch instructions in a bunch of instructions is disclosed. The method includes indicating a direction of a plurality of branch instructions in a bunch of instructions. More particularly, the method includes building an address composed of an instruction fetch address and bits in a history register. The method accesses a bunch of instructions using the fetch address and accesses a prediction bits set from a branch history table using the composed address. The accessed bunch of instructions are processed. Further, the history register and the branch history table are updated to correlate a first group of a branch instructions in the accessed bunch of instructions to a second group of branch instructions in a next group of branch instructions in the bunch of instructions.
REFERENCES:
patent: 5553253 (1996-09-01), Pan et al.
patent: 5687360 (1997-11-01), Chang
patent: 5758143 (1998-05-01), Levitan
patent: 5794063 (1998-08-01), Favor
patent: 5805878 (1998-09-01), Rahman et al.
patent: 5815699 (1998-09-01), Puziol et al.
patent: 5875325 (1999-02-01), Talcott
Kulkarni Paritosh M.
Reeve Richard
Saxena Nirmal R.
Fujitsu Ltd.
Kim Kenneth S.
LandOfFree
Predicting for all branch instructions in a bunch based on histo does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Predicting for all branch instructions in a bunch based on histo, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Predicting for all branch instructions in a bunch based on histo will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1002938