Reconfigurable memory with selectable error correction storage

• US 6,584,543 B2
• Filed: 11/14/2002
• Issued: 06/24/2003
• Est. Priority Date: 07/22/1999
• Status: Expired due to Term

• Current Assignee: Round Rock Research LLC
• Original Assignee: Micron Technology, Inc.
• Inventors: Baldwin, Donald D., Williams, Brett L.
• Primary Examiner(s): Thai, Tuan V.
• Application Number:
  US10/295,661
• Publication Number:
  US 20030070054A1
• Time in Patent Office: 222 Days
• Field of Search: 711/100, 711/103, 711/105, 711/154, 711/173, 714/704, 714/708, 714/48, 714/764, 717/723
• US Class Current: 711/105
• CPC Class Codes: G06F 11/1052 Bypassing or disabling error detection or correction

Abstract

A memory structure includes a memory module divided into low order banks and high order banks. The low order banks are used as conventional memory. The high order banks are used as either conventional memory or ECC memory, depending upon routing of data. In one embodiment, data from the high order banks are routed through a primary multiplexer to a data bus when the high order banks are used as conventional memory. When the high order banks are used as ECC memory, data from the auxiliary section is routed through the primary multiplexer to an error correction circuit. A secondary multiplexer combines ECC bits from the auxiliary section of the module or a dedicated ECC memory on a motherboard. The auxiliary section thus supplements the onboard ECC memory to provide support for an effectively larger ECC memory for use with error intolerant applications that require error correction.

Claims

1. A method of storing data in a memory device having first and second memory portions, comprising:

selecting either an error correction mode or a non-error correction mode;

storing data in the first memory portion;

if the error correction mode is selected, storing error correction bits corresponding to the data in the second memory portion; and

if the non-error correction mode is selected, storing information data in the second memory portion, the information data being comprised of other than error correction data or error detection data.

2. The method of claim 1 further comprising:

retrieving data from the first memory portion;

if the error correction mode is selected, retrieving error correction bits corresponding to the data from the second memory portion; and

if the non-error correction mode is selected, retrieving data from the second memory portion.

3. The method of claim 2 further comprising, if the error correction mode is selected:

determining from the error correction bits if the retrieved data is in error; and

if the retrieved data is determined to be in error, correcting the retrieved data using the error correction bits.

4. The method of claim 1 wherein the storing of data in the second memory portion comprises:

storing a first segment of a data byte in a first location having a first address in the second memory portion; and

storing a second segment of the data byte in a second location having a second address in the second memory portion.

5. The method of claim 4, further comprising:

retrieving the stored first segment from the first location at a first time;

retrieving the stored second segment from the second location at a second time different from the first time; and

combining the retrieved first and second segments.

6. A method of retrieving data from a memory device having first and second memory portions, comprising:

selecting either an error correction mode or a non-error correction mode;

retrieving data from the first memory portion;

if the error correction mode is selected, retrieving error correction bits corresponding to the data from the second memory portion; and

if the non-error correction mode is selected, retrieving information data from the second memory portion, the data retrieved being comprised of other than error correction data or error detection data.

7. The method of claim 6 further comprising, if the error correction mode is selected:

determining from the error correction bits if the retrieved data is in error; and

if the retrieved data is determined to be in error, correcting the retrieved data using the error correction bits.