Browse Prior Art Database

Sense Information and Control for Storage Controller

IP.com Disclosure Number: IPCOM000077349D
Original Publication Date: 1972-Jul-01
Included in the Prior Art Database: 2005-Feb-25
Document File: 1 page(s) / 12K

Publishing Venue

IBM

Related People

Hancock, RJ: AUTHOR [+2]

Abstract

Described is a method of structuring and encoding sense information between a storage control and a CPU for allowing efficient identification of error type, and for variable construction of detail error information in case of an abnormal condition in subsystem operation. A sense control block is implemented which defines: 1) the type of error which occurs; 2) the construction of detailed error information; 3) the location of the sense information; 4) the restart command. This information is contained in sense information bytes which are transmitted from the controller to the CPU.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 77% of the total text.

Page 1 of 1

Sense Information and Control for Storage Controller

Described is a method of structuring and encoding sense information between a storage control and a CPU for allowing efficient identification of error type, and for variable construction of detail error information in case of an abnormal condition in subsystem operation. A sense control block is implemented which defines: 1) the type of error which occurs; 2) the construction of detailed error information; 3) the location of the sense information; 4) the restart command. This information is contained in sense information bytes which are transmitted from the controller to the CPU.

The type of error which occurred is decoded in predetermined sense bytes. These bytes are then used by the CPU to identify the error-recovery action to be taken for the type of error condition which occurred. This type of error information can be initialized at the time the error is detected. If the error becomes permanent this information can be updated to reflect this change.

The construction of the detailed error information can be implemented as format code and message code. These two types of code can be contained in a given one of the sense bytes. For example, the format code could be encoded in a predetermined number of bits of the byte and will define the construction of subsequent sense bytes transmitted from the controller to the CPU. The message code is encoded in other bits of the aforementioned byte and used to provide fur...