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

Concurrent Memory Diagnostics

IP.com Disclosure Number: IPCOM000076116D
Original Publication Date: 1972-Jan-01
Included in the Prior Art Database: 2005-Feb-24
Document File: 2 page(s) / 30K

Publishing Venue

IBM

Related People

Harper, DR: AUTHOR [+2]

Abstract

The concurrent array diagnostic program is a task which can isolate array failures to the field replaceable unit in an environment with other tasks and I/O.

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 72% of the total text.

Page 1 of 2

Concurrent Memory Diagnostics

The concurrent array diagnostic program is a task which can isolate array failures to the field replaceable unit in an environment with other tasks and I/O.

Concurrent diagnostics is accomplished through the use of integrated diagnostic hardware in the Storage Distribution Element (SDE) and software. When an uncorrectable error occurs, the memory diagnostic program is run as a part of the job stream. Suggested up-to-date repair action is printed and placed in a log. Communications between the processor and the memory is attained by using one program Event Recording Register (say control register #10) and one maintenance control word (MCW) bit (see Fig. 1). The MCW bit is used to sensitize control logic in the SDE, so that a memory address range and a memory diagnostic instruction code may be transferred from the Program Event Recording register to a register inside the SDE. As memory requests are made, the address in the SDE register is compared with the address on the bus. If the address on the bus falls within address range, the operation specified by the instruction is performed. The special memory operations are as follows: 1. Fetch Data Bits.

Data bits with parity are sent to the CPU. The bits are sent uncorrected (error-correction code (ECC) by-passed). The data address is the address sent to the SDE from the CPU address bus. The CPU would then place the data in some Floating Point register to be analyzed. 2) Fetch Check Bits....