Modifying data prefetching operation based on a past prefetching attempt

• US 8,954,680 B2
• Filed: 11/20/2011
• Issued: 02/10/2015
• Est. Priority Date: 11/20/2011
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Peleg, Nitzan
• Primary Examiner(s): Bansal, Gurtej
• Application Number:
  US13/300,587
• Publication Number:
  US 20130132680A1
• Time in Patent Office: 1,178 Days
• Field of Search: 711/133, 711/137
• US Class Current: 711/137
• CPC Class Codes: G06F 12/0862 with prefetch

Abstract

A method, apparatus and product for data prefetching. The method comprising: prefetching data associated with a load instruction of a computer program, wherein the prefetching is performed in anticipation to performing the load instruction, whereby the data is retained in the cache; detecting whether the prefetched data of the prefetching is invalidated after the prefetching commenced and prior to performing the load instruction; and adaptively determining whether to modify the prefetching data operation associated with the load instruction in response to the detection.

Claims

1. A computer-implemented method performed by a processor having a transactional memory and a cache, the method comprising:

initiating a conflict detection transaction in the transactional memory, wherein the conflict detection transaction is configured to detect a read-write conflict in the transactional memory, wherein the conflict detection transaction is configured to perform prefetching of data associated with a load instruction of a computer program, wherein said prefetching is performed in anticipation to performing the load instruction, whereby the data is retained in the cache, wherein the conflict detection transaction is configured to end immediately upon completion of said prefetching;

in response to a modification of the data that is loaded during the conflict detection transaction, the transactional memory detecting a read-write conflict; and

in response to said detecting, adaptively determining whether to modify the prefetching data operation associated with the load instruction thereby treating future operations of the load instruction differently.

2. The computer-implemented method of claim 1, wherein said prefetching is performed by a processing entity and wherein the load instruction is performed by a different processing entity.

3. The computer-implemented method of claim 1, wherein the load instruction is a delinquent load instruction of the computer program.

4. The computer-implemented method of claim 1, wherein the load instruction is an instruction within a critical section within the computer program.

5. The computer-implemented method of claim 1, wherein the determination by said adaptively determining is stored and reused in future executions of the computer program by said processor or processors having similar configurations thereof.

6. The computer-implemented method of claim 1, wherein said adaptively determining comprises reducing expected time difference between said prefetching and performing of the load instruction.

7. The computer-implemented method of claim 1, wherein said prefetching is performed in anticipation to performing the load instruction comprises performing the prefetching at a determined timing prior to an expected timing of performing the load instruction, wherein the determined timing is about an expected cache miss latency associated with fetching the data.

8. The computer-implemented method of claim 1,

wherein the load instruction is configured to load data from a target address;

wherein said prefetching comprises;

computing a first address based on the target address; and

prefetching the data from the first address; and

wherein the invalidation is a modification that affects the computation of the target address, whereby after the modification the load instruction is configured to load data from a second address.

9. The computer-implemented method of claim 1, wherein said adaptively determining comprises:

determining whether to avoid activating the prefetching data operation associated with the load instruction.

10. The computer-implemented method of claim 9, wherein said adaptively determining comprises:

calculating a ratio between a number of successful prefetching operations and unsuccessful prefetching operations, and determining to avoid prefetching data associated with the load instruction in response to the ratio being below a threshold.

11. The computer-implemented method of claim 9, wherein said adaptively determining comprises:

calculating a ratio between a cumulative estimated performance speedup of successful prefetching operations and cumulative estimated performance reductions of unsuccessful prefetching operations, and determining to avoid prefetching data associated with the load instruction in response to the ratio being below a threshold.

12. A computerized apparatus comprising:

a transactional memory for retaining data, wherein said transactional memory is operatively coupled to a cache; and

a processor, having an associated cache, which is arranged to;

initiate a conflict detection transaction in said transactional memory, wherein the conflict detection transaction is configured to detect a read-write conflict in said transactional memory, wherein the conflict detection transaction is configured to prefetch data associated with a load instruction of a computer program, wherein the prefetching is performed in anticipation to performing the load instruction, whereby the data is retained in the cache, wherein the conflict detection transaction is configured to end immediately upon completion of said prefetching;

in response to a modification of the data that is loaded during the conflict detection transaction, the transactional memory is configured to detect a read-write conflict; and

in response to the detection, adaptively determine whether to modify the prefetching data operation associated with the load instruction thereby treating future operations of the load instruction differently.

13. The computerized apparatus of claim 12, wherein said processor is operative to store the adaptive determination and reuse the adaptive determination in future executions of the computer program.

14. The computerized apparatus of claim 12, wherein the adaptive determination is a determination whether to avoid activating the prefetching data operation associated with the load instruction.

15. The computerized apparatus of claim 14, wherein the adaptive determination comprises:

calculating a ratio between a number of successful prefetching operations and unsuccessful prefetching operations, and determining to avoid prefetching data associated with the load instruction in response to the ratio being below a threshold.

16. A computer program product comprising:

a non-transitory computer readable medium;

a first program instruction for initiating a conflict detection transaction in a transactional memory, wherein the conflict detection transaction is configured to detect a read-write conflict in the transactional memory, wherein the conflict detection transaction is configured to perform prefetching of data associated with a load instruction of a computer program, wherein said prefetching is performed in anticipation to performing the load instruction, whereby the data is retained in the cache, wherein the conflict detection transaction is configured to end immediately upon completion of said prefetching;

a second program instruction for detecting, by the transactional memory and in response to a modification of the data that is loaded during the conflict detection transaction, a read-write conflict;

a third program instruction for adaptively determining, in response to said detecting, whether to modify the prefetching data operation associated with the load instruction in response to said detection thereby treating future operations of the load instruction differently; and

wherein said first, second, and third program instructions are stored on said non-transitory computer readable medium.