Browse Prior Art Database

System Detection of Inoperative Service Processor

IP.com Disclosure Number: IPCOM000041674D
Original Publication Date: 1984-Feb-01
Included in the Prior Art Database: 2005-Feb-02
Document File: 1 page(s) / 11K

Publishing Venue

IBM

Related People

Bullions, RJ: AUTHOR [+6]

Abstract

In today's environment a service processor will perform many functions for a system it is supporting. One of them is to detect if the central processor (CP) or channel engine is stalled. In addition, as described below, the CP and channel can detect if the Service Processor is stalled. If the Service Processor is stalled, the CP generates a machine check interrupt (MCI) with a special MCI condition bit and sets an indicator. The system control program (SCP) can then gracefully shut down the system and, when the system starts again, can recover using the data that is stored. The mechanism to detect if the service processor is stalled uses the CP to set a bit in storage and send a signal to the service processor. In a predetermined amount of time, the CP checks to see if the Service Processor has reset the bit in storage.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 100% of the total text.

Page 1 of 1

System Detection of Inoperative Service Processor

In today's environment a service processor will perform many functions for a system it is supporting. One of them is to detect if the central processor (CP) or channel engine is stalled. In addition, as described below, the CP and channel can detect if the Service Processor is stalled. If the Service Processor is stalled, the CP generates a machine check interrupt (MCI) with a special MCI condition bit and sets an indicator. The system control program (SCP) can then gracefully shut down the system and, when the system starts again, can recover using the data that is stored. The mechanism to detect if the service processor is stalled uses the CP to set a bit in storage and send a signal to the service processor. In a predetermined amount of time, the CP checks to see if the Service Processor has reset the bit in storage. If it has, the Service Processor is not stalled. If the bit is not reset, the CP will set up an MCI in all CPs. The channels, before calling for Service Processor service such as collecting data for interface control checks, test the storage bit to see if the Service Processor; is stalled as set by the detecting CP. If yes, don't signal the Service Processor, if no, then signal as normally. In addition special instructions like SIG P that look for Service Processor support, also test the Service Processor stalled bit. If on, it will set a special condition code.

1