Fault-tolerant method and means for managing access to an initial program load stored in read-only memory or the like
US5832005A · kind A · utility
Assignee
Inventor
Key dates
| Filing date | Dec 11, 1997 |
| Grant date | Nov 3, 1998 |
| Priority date | — |
| Expiry date | Dec 11, 2017 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06F11/1417
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A method and means in a stored, program-controlled machine to maintain an uninterrupted access and copying of simple parity-coded words of an initial program load (IPL) or other software from a read-only memory or a memory with limited rewrite capability in the presence of detectable errors, erasures, or faults. The method utilizes detected parity error in a word copied out from storage to modify the address register to seek the same word from a mirror address of an image copy of the IPL. The image copy is stored at a second range of consecutive memory address nonintersecting and symmetric with the address range of the original IPL. The IPL is then accessed in the second address range until either another parity error causes a switch back to the original IPL or the program terminates.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.