Managing a prefetch queue based on priority indications of prefetch requests

Patent metadata

Field	Value
Publication number	US-10169239-B2
Application number	US-201615214549-A
Country	US
Kind code	B2
Filing date	Jul 20, 2016
Priority date	Jul 20, 2016
Publication date	Jan 1, 2019
Grant date	Jan 1, 2019

Abstract

A prefetch request having a priority assigned thereto is obtained, based on executing a prefetch instruction included within a program. Based on obtaining the prefetch request, a determination is made as to whether the prefetch request may be placed on a prefetch queue. This determination includes determining whether the prefetch queue is full; checking, based on determining the prefetch queue is full, whether the priority of the prefetch request is considered a high priority; determining, based on the checking indicating the priority of the prefetch request is considered a high priority, whether another prefetch request on the prefetch queue may be removed; removing the other prefetch request from the prefetch queue, based on determining the other prefetch request may be removed; and adding the prefetch request to the prefetch queue, based on removing the other prefetch request.

First claim

What is claimed is:

1. A computer program product for managing prefetch queues, said computer program product comprising:
a computer readable storage medium readable by a processing circuit and storing instructions for execution by the processing circuit for performing a method comprising:
obtaining a prefetch request based on executing a prefetch instruction included within a program and determining that data at a memory location designated by the prefetch instruction is not located in a selected cache, the prefetch request having a priority assigned thereto, the priority indicating confidence by the program in whether the prefetch request will be used;
determining, based on obtaining the prefetch request, whether the prefetch request may be placed on a prefetch queue, the determining comprising:
determining whether the prefetch queue is full;
checking, based on determining the prefetch queue is full, whether the priority of the prefetch request is considered a high priority;
determining, based on the checking indicating the priority of the prefetch request is considered a high priority, whether another prefetch request on the prefetch queue may be removed;
removing the other prefetch request from the prefetch queue, based on determining the other prefetch request may be removed; and
adding the prefetch request to the prefetch queue, based on removing the other prefetch request.

2. The computer program product of claim 1 , wherein the other prefetch request is assigned a priority that is considered a low priority.

3. The computer program product of claim 1 , wherein the priority is selected from more than two possible priorities.

4. The computer program product of claim 1 , wherein the method further comprises discarding the prefetch request, based on the checking indicating the priority of the prefetch request is not considered a high priority.

5. The computer program product of claim 1 , wherein the method further comprises determining how to process the prefetch request, based on determining another prefetch request may not be removed.

6. The computer program product of claim 5 , wherein the determining how to process the prefetch request includes choosing to defer processing of the prefetch request for a defined time.

7. The computer program product of claim 1 , wherein the method further comprises:
determining, based on the prefetch queue not being full, whether the prefetch queue is nearly full in that it can accept a predefined number of additional requests or there is a low cache bandwidth in that an arrival rate of cache requests has a predefined relationship with respect to a number of bytes per a select number of cycles;
checking, based on determining the prefetch queue is nearly full or there is a low cache bandwidth, whether the priority of the prefetch request is considered a high priority; and
adding the prefetch request to the prefetch queue based on determining the prefetch queue is nearly full or there is a low cache bandwidth and the priority of the prefetch request is considered a high priority.

8. The computer program product of claim 7 , wherein the method further comprises discarding the prefetch request, based on determining the prefetch queue is nearly full or there is a low cache bandwidth and the priority of the prefetch request is not considered a high priority.

9. The computer program product of claim 7 , wherein the method further comprises adding the prefetch request to the prefetch queue, based on determining the prefetch queue is not nearly full or there is not a low cache bandwidth.

10. The computer program product of claim 1 , wherein the method further comprises:
determining, based on the prefetch queue not being full, whether there is a low cache bandwidth in that an arrival rate of cache requests has a predefined relationship with respect to a number of bytes per a select number of cycles;
checking, based on determining there is a low cache bandwidth, whether the priority of the prefetch request is considered a high priority; and
adding the prefetch request to the prefetch queue based on determining there is a low cache bandwidth and the priority of the prefetch request is considered a high priority.

11. The computer program product of claim 1 , wherein the selected cache is a level 1 cache.

12. A computer system for managing prefetch queues, said computer system comprising:
a memory; and
a processor in communication with the memory, wherein the computer system is configured to perform a method, said method comprising:
obtaining a prefetch request based on executing a prefetch instruction included within a program and determining that data at a memory location designated by the prefetch instruction is not located in a selected cache, the prefetch request having a priority assigned thereto, the priority indicating confidence by the program in whether the prefetch request will be used;
determining, based on obtaining the prefetch request, whether the prefetch request may be placed on a prefetch queue, the determining comprising:
determining whether the prefetch queue is full;
checking, based on determining the prefetch queue is full, whether the priority of the prefetch request is considered a high priority;
determining, based on the checking indicating the priority of the prefetch request is considered a high priority, whether another prefetch request on the prefetch queue may be removed;
removing the other prefetch request from the prefetch queue, based on determining the other prefetch request may be removed; and
adding the prefetch request to the prefetch queue, based on removing the other prefetch request.

13. The computer system of claim 11 , wherein the method further comprises:
determining, based on the prefetch queue not being full, whether the prefetch queue is nearly full in that it can accept a predefined number of additional requests or there is a low cache bandwidth in that an arrival rate of cache requests has a predefined relationship with respect to a number of bytes per a select number of cycles;
checking, based on determining the prefetch queue is nearly full or there is a low cache bandwidth, whether the priority of the prefetch request is considered a high priority; and
adding the prefetch request to the prefetch queue based on determining the prefetch queue is nearly full or there is a low cache bandwidth and the priority of the prefetch request is considered a high priority.

14. The computer system of claim 13 , wherein the method further comprises discarding the prefetch request, based on determining the prefetch queue is nearly full or there is a low cache bandwidth and the priority of the prefetch request is not considered a high priority.

15. The computer system of claim 13 , wherein the method further comprises adding the prefetch request to the prefetch queue, based on determining the prefetch queue is not nearly full or there is not a low cache bandwidth.

16. A computer-implemented method of managing prefetch queues, said computer-implemented method comprising:
obtaining, by a processor, a prefetch request based on executing a prefetch instruction included within a program and determining that data at a memory location designated by the prefetch instruction is not located in a selected cache, the prefetch request having a priority assigned thereto, the priority indicating confidence by the program in whether the prefetch request will be used;
determining, based on obtaining the prefetch request, whether the prefetch request may be placed on a prefetch queue, the determining comprising:
determining w

Assignees

• IBM

Inventors

• Greiner Dan F
• Gschwind Michael K
• Jacobi Christian
• Saporito Anthony
• Shum Chung-Lung K
• Slegel Timothy J

Classifications

• G06F2212/602

  Details relating to cache prefetching · CPC title

• G06F12/0875

  with dedicated cache, e.g. instruction or stack · CPC title

• G06F2212/6028

  Prefetching based on hints or prefetch instructions · CPC title

• G06F12/0862 Primary

  with prefetch · CPC title

• G06F2212/1016

  Performance improvement · CPC title