[go: up one dir, main page]

WO1996007969A1 - On board error correction apparatus - Google Patents

On board error correction apparatus Download PDF

Info

Publication number
WO1996007969A1
WO1996007969A1 PCT/CA1995/000517 CA9500517W WO9607969A1 WO 1996007969 A1 WO1996007969 A1 WO 1996007969A1 CA 9500517 W CA9500517 W CA 9500517W WO 9607969 A1 WO9607969 A1 WO 9607969A1
Authority
WO
WIPO (PCT)
Prior art keywords
memory
computer
rsimc
parity
eca
Prior art date
Application number
PCT/CA1995/000517
Other languages
French (fr)
Inventor
Bosco C. S. Lai
Original Assignee
Lai Bosco C S
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from CA 2131735 external-priority patent/CA2131735A1/en
Application filed by Lai Bosco C S filed Critical Lai Bosco C S
Publication of WO1996007969A1 publication Critical patent/WO1996007969A1/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/08Error detection or correction by redundancy in data representation, e.g. by using checking codes
    • G06F11/10Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's
    • G06F11/1008Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's in individual solid state devices
    • G06F11/1044Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's in individual solid state devices with specific ECC/EDC distribution
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C29/00Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/70Masking faults in memories by using spares or by reconfiguring
    • G11C29/88Masking faults in memories by using spares or by reconfiguring with partially good memories

Definitions

  • the present invention is an on board errors correction methodology whereas an error correction apparatus (ECA) is a separate chip or device which utilizes the error detection and correction method in a memory system and memory sub-systems.
  • An algorithm detects and reports the kind of errors of the RSIMC (Memory chip that does not meet the manufacturers' original designed specifications). RSIMC with the same error pattern will then be grouped together along with the ECA for assemble into memory modules or memory cards .
  • ECA employs the redundant Error Correcting Algorithm which enables the rejected, fallout, audio grade, toy grade or any other kinds of reduced specification memory chips to become computer grade compatible, so that the corrected RSIMC memory modules would function as normal memory modules for computer (or computer sub-systems) usages and applications.
  • the invention is an "on board" error correction methodology which employs the Redundant Error Correcting Algorithm to upgrade the RSIMC to become computer grade compatible.
  • ECA is mounted on the memory boards, memory cards, or SIMM modules along with the RSIMC's. It corrects all kinds of RSIMC errors as long as there is only one single defect in the same memory address at a time. There is no limit on the number of errors corrected in total.
  • the corrected RSIMC memory modules would function as normal memory modules for computer (or computer sub-systems) usages and applications and would only have a slight adverse effect on the computers' speed performance.
  • RSIMC refers to Reduced Specification Integrated Memory Chip which does not meet the manufacturers' original designed specifications. It includes but not limited to DRAM, ARAM, SRAM, VRAM, ROM, PROM, EPROM, FRESH Memory, and EEPROM.
  • On Board ECC means the ECC is done within the memory cards, memory modules, SIMM, or memory subsystems.
  • the invention is an "on board" error correction methodology which employs the Redundant Error Correcting Algorithm to upgrade the RSIMC to make the RSIMC board, RSIMC card or RSIMC module to become computer grade compatible.
  • An algorithm detects and reports the kind of errors of the RSIMC.
  • RSIMC with the same error pattern will then be grouped together along with the ECA for assembly into memory boards, memory cards or SIMM modules.
  • ECA is a redundant backup / switch method which corrects all kinds of RSIMC errors as long as there is only one single defect in the same memory address at a time.
  • the corrected RSIMC memory board, memory cards or SIMM modules would function as normal memories for computer (or computer sub-systems) usages and applications. It will only have a slight adverse effect on the computers' speed performance.
  • the ECA includes the following components:
  • This part generates an extra 4 check bits Cl, C2, C3 and C4 for parity check.
  • the followings show an example of how the algorithm can be implemented.
  • This part generates the expecting check bits and compares the bits with Cl, C2, C3 and C4 using the same algorithm as in part 1.
  • the compared result will indicate the location of error if there are any errors occurred during the checking process.
  • the compared result will provide instructions to the correction circuit to correct the errors accordingly.
  • This part is the logic circuit to control the input / output timing for the ECA. It enables the output driver only after the output value is fixed and stabilized. It also ensures the , latch / input driver to be compatible with different speed devices. In other words, it improves the turn on/off lead time.
  • This part is to capture and to generate the parity mode (Odd / Even) of the system by using the "x 8 configuration”. It will then generate one of the following messages:
  • This part is to control input and output information of the memory chips, and to latch data. It isolates the memory chips and the data bus by acting as a I/O buffer to the memory chips. Therefore, instead of communicating to the data bus, the memory chips only "talk" to the ECA. Thus, it releases the I/O drivers' requirement for the memory chips.
  • the input latch also makes a clear input speed limit so that the ECA will not be affected by any unstable inputs.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Techniques For Improving Reliability Of Storages (AREA)

Abstract

This invention includes an algorithm, which detects and identifies the error in each memory address of the Reduced Specification Integrated Memory chips (RSIMC). RSIMC with the same error pattern will then be grouped together along with the Error Correction Apparatus (ECA) for assembly into memory modules. ECA employs the redundant Error Correcting Algorithm which enables the rejected, fallout, audio grade, toy grade and other reduced specification memory chips to become computer grade compatible, such that the corrected RSIMC memory modules would function as normal memory modules for computer (or computer sub-systems) usages and applications.

Description

TITLE
ON BOARD ERROR CORRECTION APPARATUS
Background of the Invention
Field of the Invention:
The present invention is an on board errors correction methodology whereas an error correction apparatus (ECA) is a separate chip or device which utilizes the error detection and correction method in a memory system and memory sub-systems. An algorithm detects and reports the kind of errors of the RSIMC (Memory chip that does not meet the manufacturers' original designed specifications). RSIMC with the same error pattern will then be grouped together along with the ECA for assemble into memory modules or memory cards . ECA employs the redundant Error Correcting Algorithm which enables the rejected, fallout, audio grade, toy grade or any other kinds of reduced specification memory chips to become computer grade compatible, so that the corrected RSIMC memory modules would function as normal memory modules for computer (or computer sub-systems) usages and applications.
Description of the Prior Act.
The U.S. patent specification No. 5,127,014 discloses an error detection or correction method on the same chip as DRAM memory.
The invention of the Prior Act was an "On Chip ECC" to reduce the reject rate of the memory chips, not an "On Board ECC". Summary of the Invention
It is an object of the invention to utilize the RSIMC in computer applications. The invention is an "on board" error correction methodology which employs the Redundant Error Correcting Algorithm to upgrade the RSIMC to become computer grade compatible. ECA is mounted on the memory boards, memory cards, or SIMM modules along with the RSIMC's. It corrects all kinds of RSIMC errors as long as there is only one single defect in the same memory address at a time. There is no limit on the number of errors corrected in total.
The corrected RSIMC memory modules would function as normal memory modules for computer (or computer sub-systems) usages and applications and would only have a slight adverse effect on the computers' speed performance.
RSIMC refers to Reduced Specification Integrated Memory Chip which does not meet the manufacturers' original designed specifications. It includes but not limited to DRAM, ARAM, SRAM, VRAM, ROM, PROM, EPROM, FRESH Memory, and EEPROM.
"On Board" ECC means the ECC is done within the memory cards, memory modules, SIMM, or memory subsystems.
Detailed Description of the Invention
The invention is an "on board" error correction methodology which employs the Redundant Error Correcting Algorithm to upgrade the RSIMC to make the RSIMC board, RSIMC card or RSIMC module to become computer grade compatible.
An algorithm detects and reports the kind of errors of the RSIMC. RSIMC with the same error pattern will then be grouped together along with the ECA for assembly into memory boards, memory cards or SIMM modules.
ECA is a redundant backup / switch method which corrects all kinds of RSIMC errors as long as there is only one single defect in the same memory address at a time. The corrected RSIMC memory board, memory cards or SIMM modules would function as normal memories for computer (or computer sub-systems) usages and applications. It will only have a slight adverse effect on the computers' speed performance.
The ECA includes the following components:
1. Check Bit generating circuit for encoding purpose in the memory write cycle:
This part generates an extra 4 check bits Cl, C2, C3 and C4 for parity check. The followings show an example of how the algorithm can be implemented.
C1 =DlEORD2EORD4EORD5EORD7
C2=DlEORD3EORD4EORD6EORD7
C3=D2EORD3EORD4EORD8
C4=D5EORD6EORD7EORD8
2. Check Bit generating circuit for decoding purpose in the memory read cycle:
This part generates the expecting check bits and compares the bits with Cl, C2, C3 and C4 using the same algorithm as in part 1. The compared result will indicate the location of error if there are any errors occurred during the checking process. The compared result will provide instructions to the correction circuit to correct the errors accordingly. 3. Timing control circuit
This part is the logic circuit to control the input / output timing for the ECA. It enables the output driver only after the output value is fixed and stabilized. It also ensures the , latch / input driver to be compatible with different speed devices. In other words, it improves the turn on/off lead time.
4. Parity capture / Generating circuit
This part is to capture and to generate the parity mode (Odd / Even) of the system by using the "x 8 configuration". It will then generate one of the following messages:
(a) Same parity message if there is no error or if there is only one correctable error detected;
(b) Different parity message if there is non-correctable error(s). In this case, it will generat parity error message to the computer system.
Since this "x 8 configuration" generates the parity mode itself, it is compatible with other systems using the parity feature.
5. I/O circuit for memory function
This part is to control input and output information of the memory chips, and to latch data. It isolates the memory chips and the data bus by acting as a I/O buffer to the memory chips. Therefore, instead of communicating to the data bus, the memory chips only "talk" to the ECA. Thus, it releases the I/O drivers' requirement for the memory chips.
The input latch also makes a clear input speed limit so that the ECA will not be affected by any unstable inputs.

Claims

ClaimsI claim:
1. An "on board" error correction methodology which employs the Redundant Error Correction Algorithm to upgrade the RSIMC's to make the RSIMC boards, RSIMC cards or RSIMC modules to become computer grade compatible. Such that the corrected RSIMC's could be applied on computer's main memory system or memory subsystem.
2. The methodology, as claimed in claim 1, includes an error detection algorithm which reports the kind of errors of the RSIMC.
3. RSIMC, as claimed in claim 1 and 2, includes all kinds of reduced specification integrated memory chips which do not meet the manufacturers' original designed specification. Names such as rejected chips, fallout chips, audio grade chips or toy grade chips are also considered as RSIMC.
4. Memories or Memory chips refer to the data width of the folio wings but not limited to:
(a) 30 pins, x8 or x9 configuration SIMM;
(b) 72 pins, x32, x36 or x40 configuration SIMM;
(c) 60 pins, xl6 or xl8 configuration SIMM;
(d) other x64, or xl28 configuration SIMM; and
(d) PCMCIA memory cards, DRAM, ARAM, SRAM, VRAM, ROM, PROM, EPROM, FRESH Memory, and EEPROM.
5. The methodology, as claimed in claim 1, includes an apparatus which employs the Redundant Error Correction Algorithm (ECA). It has the following components:
(a) Check Bit generating circuit for encoding purpose:
This part generates an extra 4 check bits Cl, C2, C3 and C4 for parity check. The followings show an example of how the algorithm can be implemented. Cl=DlEORD2EORD4EORD5EORD7
C2=DlEORD3EORD4EORD6EORD7
C3=D2EORD3EORD4EORD8
C4=D5EORD6EORD7EORD8
(b) Check Bit generating circuit for decoding purpose:
This part generates the expecting check bits and to compare the bits with Cl, C2, C3 and C4 using the same algorithm as in part (a). The compared result will indicate the location of errors if there are any errors occurred during the checking process. It will provide instructions to the correction circuit to correct the errors accordingly.
(c) Timing control circuit
This part is the logic circuit to control the input / output timing for the ECA. It enables the output driver only after the output value is fixed and stabilized. It also ensures the latch / input driver to be compatible with different speed devices. In other words, it improves the turn on off lead time.
(d) Parity capture / Generating circuit
This part is to capture and to generate the parity mode (Odd / Even) of the system by using the "x 8 configuration". It will then generate one of the following messages: (i) Same parity message if there is no error or if there is only one correctable error detected; (ii) Different parity message if there is non-correctable error(s). In this case, it will generate parity error message to the computer system.
Since this "x 8 configuration" generates the parity mode itself, it is compatible with other systems using the parity feature.
The ECC circuit detects and corrects all single error as it occurs. It also generates the parity check information according to the system's parity mode.
- 6 -
SUBSTTTUTE SHEET (RULE 26) (e) I/O circuit for memory function
This part is to control input and output information of the memory chips, and to latch data. It isolates the memory chips and the data bus by acting as a I/O buffer to the memory chips. Therefore, instead of communicating to the data bus, the memory chips only "talk" to the ECA. Thus, it releases the I/O drivers' requirement for the memory chips. The input latch also makes a clear input speed limit so that the ECA will not be affected by any unstable inputs.
6. The components, as claimed in claim 5, have the following features:
(a) added timing control to input and output driver to ensure its' compatibility with different speed devices.
(b) changed the data input level from TTL to CMOS to add noise margins.
(c) added learning circuit to the parity generating circuit to make sure its compatibility to both ODD and EVEN parity check.
(d) made 'CAS' before *RAS', Hidden' Refresh compatible.
(e) added input latch to make a clear input speed limit so that the ECA won't be affected by any unstable input signals.
(f) modified parity check circuit such that it generates parity error if the ECA failed.
(g) modified the internal control logic to improve its turn on/off lead time.
7. The methodology, as claimed in claim 1, whereas the RSIMC with the same error pattern will then be grouped together along with the ECA for assembly into memory boards, memory cards or SIMM modules.
8. The methodology, as claimed in claim 1, whereas the ECA is a redundant backup / switch method which corrects all kinds of RSIMC errors as long as there is only one single defect in the same memory address at a time.
9. The methodology, as claimed in claim 1, whereas the corrected RSIMC memory board, memory cards or SIMM modules would become computer grade compatible and would function as normal memories for computer (or computer sub-systems) usages and applications. It will only have a slight adverse effect on the computers' speed performance.
10. Computers), as claimed in claim 9, include(s) any device with a central processing unit, an input system, output system, memory system and a data bus connecting the central processing unit to the memory system.
11. Redundant Error Detection / Correction Algorithm, as claimed in claim 1, refers to any kinds of error correction method, procedure or algorithm. They include but not limited to HAMING CODE, OR MODIFY HAMING CODE.
12. The methodology, as claimed in claim 1, is the integration of any the ECC technology on computer's memory or memory sub-system for computers) with or without built in ECC capacity.
13. The methodology, as claimed in claim 1, is the integration of any the ECC technology on a computer's memory cards, memory boards, or SIMM modules such that the host computer will access the memory cards, memory boards, or SIMM modules as good quality memory chips. There is no limit on the address size of the memory chip/ card/ module or SIMM that it can correct.
14. On Board ECC as claimed in claim 1, means the ECC is done within the memory cards, memory modules, SIMM, or memory subsystems for the purpose of upgrading the reduced specification memory chips to become computer grade compatible such that the corrected memory cards, modules or SIMM can be used in a computer environment.
15. Built-in ECC, as mentioned in claim 12, refers to the computer with ECC capacity built inside the computer main-board for the purpose of increasing or enhancing the reliability of the memory access. To be more specific, built-in ECC is for fault-tolerance purpose.
16. The implementation of the methodology as in claim 1, will effectively replace the good quality memory card, memory board or SIMM modules with the Reduced Specification Memory Chips and the ECA such that the only effect on the computer is a slight adverse effect on the computer's speed performance but will not lower the fault-tolerance level of the computer system.
17. The purpose of the methodology, as claimed in claim 1, is to upgrade and to utilize the Reduced Specification Memory Chips in computer applications, not to increase the reliability of a computer system.
PCT/CA1995/000517 1994-09-09 1995-09-08 On board error correction apparatus WO1996007969A1 (en)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US30340094A 1994-09-09 1994-09-09
CA2,131,735 1994-09-09
CA 2131735 CA2131735A1 (en) 1994-09-09 1994-09-09 Board errors correction apparatus
US08/303,400 1994-09-09

Publications (1)

Publication Number Publication Date
WO1996007969A1 true WO1996007969A1 (en) 1996-03-14

Family

ID=25677477

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CA1995/000517 WO1996007969A1 (en) 1994-09-09 1995-09-08 On board error correction apparatus

Country Status (1)

Country Link
WO (1) WO1996007969A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0824237A3 (en) * 1996-08-05 1999-10-06 Texas Instruments Inc. Audio RAM having error detection and correction function

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3814922A (en) * 1972-12-01 1974-06-04 Honeywell Inf Systems Availability and diagnostic apparatus for memory modules
US3897626A (en) * 1971-06-25 1975-08-05 Ibm Method of manufacturing a full capacity monolithic memory utilizing defective storage cells
US4612640A (en) * 1984-02-21 1986-09-16 Seeq Technology, Inc. Error checking and correction circuitry for use with an electrically-programmable and electrically-erasable memory array
EP0520676A2 (en) * 1991-06-28 1992-12-30 STMicroelectronics, Inc. Memory subsystem with error correction
US5355377A (en) * 1993-11-23 1994-10-11 Tetra Assoc. Inc. Auto-selectable self-parity generator

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3897626A (en) * 1971-06-25 1975-08-05 Ibm Method of manufacturing a full capacity monolithic memory utilizing defective storage cells
US3814922A (en) * 1972-12-01 1974-06-04 Honeywell Inf Systems Availability and diagnostic apparatus for memory modules
US4612640A (en) * 1984-02-21 1986-09-16 Seeq Technology, Inc. Error checking and correction circuitry for use with an electrically-programmable and electrically-erasable memory array
EP0520676A2 (en) * 1991-06-28 1992-12-30 STMicroelectronics, Inc. Memory subsystem with error correction
US5355377A (en) * 1993-11-23 1994-10-11 Tetra Assoc. Inc. Auto-selectable self-parity generator

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
"Swapping Failing Bits in 40-Pin Error Correcting SIMMS", IBM TECHNICAL DISCLOSURE BULLETIN, vol. 37, no. 9, September 1994 (1994-09-01), NEW YORK US, pages 595 - 596 *
R. NASS: "MEMORY-EXPANSION SIMMS AND ERROR CORRECTION", ELECTRONIC DESIGN, vol. 42, no. 5, 7 March 1994 (1994-03-07), HASBROUCK HEIGHTS, NEW JERSEY, USA, pages 40, XP000441321 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0824237A3 (en) * 1996-08-05 1999-10-06 Texas Instruments Inc. Audio RAM having error detection and correction function

Similar Documents

Publication Publication Date Title
US5452429A (en) Error correction code on add-on cards for writing portions of data words
US5488691A (en) Memory card, computer system and method of operation for differentiating the use of read-modify-write cycles in operating and initializaiton modes
US10169126B2 (en) Memory module, memory controller and systems responsive to memory chip read fail information and related methods of operation
US7487428B2 (en) Fully-buffered memory-module with error-correction code (ECC) controller in serializing advanced-memory buffer (AMB) that is transparent to motherboard memory controller
US5923682A (en) Error correction chip for memory applications
US5623506A (en) Method and structure for providing error correction code within a system having SIMMs
CA1290459C (en) Memory system and related error detection and correction apparatus
US5379304A (en) Method and structure for providing error correction code and parity for each byte on SIMM's
US7107493B2 (en) System and method for testing for memory errors in a computer system
US20140325315A1 (en) Memory module buffer data storage
WO2008115400A1 (en) Ecc implementation in non-ecc components
US5465262A (en) Method and structure for providing error correction code and automatic parity sensing
CN1833289A (en) Hub module for connecting one or more memory chips
US20040237001A1 (en) Memory integrated circuit including an error detection mechanism for detecting errors in address and control signals
US6567950B1 (en) Dynamically replacing a failed chip
US5450422A (en) Method and structure for providing error correction code for each byte on SIMM'S
US20250060913A1 (en) Memory device, memory module including the memory device, and operating method of memory controller
US8219860B2 (en) Microprocessor system for controlling at least partly safety-critical processes
US20030182611A1 (en) Method for verifying error correction code function of a computer system
US5541941A (en) Method and structure for providing automatic parity sensing
US20040003165A1 (en) Memory subsystem including error correction
WO1996007969A1 (en) On board error correction apparatus
US20090037764A1 (en) Memory-module controller, memory controller and corresponding memory arrangement, and also method for error correction
US20100235680A1 (en) Microprocessor System for Controlling or Regulating at least partly Safety-Critical Processes
CA2131735A1 (en) Board errors correction apparatus

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): JP

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AT BE CH DE DK ES FR GB GR IE IT LU MC NL PT SE

121 Ep: the epo has been informed by wipo that ep was designated in this application
122 Ep: pct application non-entry in european phase