Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Task Switching of Special Tasks Upon Page Faults

IP.com Disclosure Number: IPCOM000042735D
Original Publication Date: 1984-Jun-01
Included in the Prior Art Database: 2005-Feb-04
Document File: 2 page(s) / 38K

Publishing Venue

IBM

Related People

Liu, L: AUTHOR [+2]

Abstract

This article discloses an algorithm by which switches to special tasks are performed upon page faults in order to utilize the page access time and retain most of the working cache lines of the faulted tasks. C is a prechosen class of special tasks that are candidates for the task-switch targets. The basic criterion for the selection of C is to choose those available tasks that do not cause much disruption to the cache contents so that the faulted task will not experience many extra cache misses. These special tasks can be chosen based on their nature and length. SRB (Service Request Block) and CMS (Conversational Monitor System) tasks existing in current MVS (Multiple Virtual Storage) systems or other similar tasks could be covered by C. Upon completion of the switched special task, the old task resumes.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 100% of the total text.

Page 1 of 2

Task Switching of Special Tasks Upon Page Faults

This article discloses an algorithm by which switches to special tasks are performed upon page faults in order to utilize the page access time and retain most of the working cache lines of the faulted tasks. C is a prechosen class of special tasks that are candidates for the task-switch targets. The basic criterion for the selection of C is to choose those available tasks that do not cause much disruption to the cache contents so that the faulted task will not experience many extra cache misses. These special tasks can be chosen based on their nature and length. SRB (Service Request Block) and CMS (Conversational Monitor System) tasks existing in current MVS (Multiple Virtual Storage) systems or other similar tasks could be covered by C. Upon completion of the switched special task, the old task resumes. In case a switched special task causes a page fault, it can be suspended and the old task resumes when ready. The faulted special task may be put in not-ready state in appropriate queues (e.g., the regular job dispatching queue or a queue for special tasks). The waking-up process of the possibly waiting faulted task depends on the facilities available in the system to indicate the page-in status. During a page fault, whether certain interrupts should be masked also depends on the specific implementations.

1

Page 2 of 2

2

[This page contains 3 pictures or other non-text objects]