Fault tolerant memory system
View Patent ↗A method and apparatus for managing X4 or larger types of memory first receives a data word to be stored in the memory, and then generates a check datum, which is a function of the data word and a set of encode data. After storing the data word in memory, the method and apparatus use the check datum and the data word to generate a syndrome. The method and apparatus then determine if the data word in the memory is correct as a function of the syndrome.
1. A method of managing X4 or larger types of memory, the method comprising:
receiving a data word to be stored in the memory;
producing a set of encode data;
generating a 16-bit check datum as a function of the data word to be stored and the encode data;
storing the data word in the memory;
using the check datum and the data word in the memory to generate a 16-bit syndrome;
determining if the data word in the memory is correct as a function of the syndrome; and
if the data word in the memory is not correct as a function of the syndrome, correcting between two and eight non-adjacent bits in the data word.
2. The method as defined by claim 1 further including storing the check datum in memory.
3. The method as defined by claim 1 wherein generating a check datum includes applying a function to selected bits of the data word, the selected bits being selected as determined by the set of encode data.
4. The method as defined by claim 1 wherein using the check datum includes:
appending the check datum with the data word to produce revised data; and
applying a logical function to selected bits of the revised data to produce the syndrome, the selected bits being selected as determined by the set of encode data to produce the syndrome.
5. The method as defined by claim 1 wherein determining if the data word in the memory is correct includes:
comparing the syndrome to a set of error identity data; and
determining that the data word in the memory is not correct if the syndrome matches at least one datum in the set of error identity data.
6. The method as defined by claim 5 wherein the set of error identity data is generated from the set of encode data, the set of error identity data including a plurality of values assigned to each bit of the data word, the error of a given bit being corrected if the syndrome matches at least one of the plurality of values assigned to the given bit.
7. The method as defined by claim 1 wherein the set of encode data includes a table identifying bits of the data word to be processed.
8. The method as defined by claim 1 further comprising:
determining that a given bit of the data word has an error; and
re-reading the given bit to determine if the error has been corrected.
9. The method as defined by claim 8 further comprising:
storing the data word in a mirror memory location; and
if the given bit is determined to have the error after it is re-read, then retrieving the correct data for the given bit from the mirror memory location and storing that corrected data in the given bit.
10. The method as defined by claim 8 further including correcting other bit errors in the data word if the given bit is corrected.
11. A computer program product for use on a computer system for managing X4 or larger types of memory, the computer program product comprising a tangible computer usable medium having computer readable program code thereon, the computer readable program code comprising:
program code for receiving a data word to be stored in the memory;
program code for producing a set of encode data;
program code for generating a 16-bit check datum as a function of the data word to be stored and the encode data;
program code for storing the data word in the memory;
program code for using the check datum and the data word in the memory to generate a 16-bit syndrome;
program code for determining if the data word in the memory is correct as a function of the syndrome; and
program code for correcting between two and eight non-adjacent bits in the data word if the data word in the memory is not correct as a function of the syndrome.
12. The computer program product as defined by claim 11 further including program code for storing the check datum in memory.
13. The computer program product as defined by claim 11 wherein the program code for generating a check datum includes program code for applying a function to selected bits of the data word, the selected bits being selected as determined by the set of encode data.
14. The computer program product as defined by claim 11 wherein the program code for using the check datum includes:
program code for appending the check datum with the data word to produce revised data; and
program code for applying a logical function to selected bits of the revised data to produce the syndrome, the selected bits being selected as determined by the set of encode data to produce the syndrome.
15. The computer program product as defined by claim 11 wherein the program code for determining if the data word in memory is correct includes:
program code for comparing the syndrome to a set of error identity data; and
program code for determining that the data word in the memory is not correct if the syndrome matches at least one datum in the set of error identity data.
16. The computer program product as defined by claim 15 wherein the set of error identity data is generated from the set of encode data, the set of error identity data including a plurality of values assigned to each bit of the data word, the error of a given bit being corrected if the syndrome matches at least one of the plurality of values assigned to the given bit.
17. The computer program product as defined by claim 11 wherein the set of encode data includes a table identifying bits of the data word to be processed.
18. The computer program product as defined by claim 11 further comprising:
program code for determining that a given bit of the data word has an error; and
program code for re-reading the given bit to determine if the error has been corrected.
19. The computer program product as defined by claim 18 further comprising:
program code for storing the data word in a mirror memory location; and
program code for retrieving the correct data for the given bit from the mirror memory location and storing that corrected data in the given bit.
20. The computer program product as defined by claim 18 further including program code for correcting other bit errors in the data word if the given bit is corrected.
21. An apparatus for managing X4 or larger types of memory, the apparatus comprising:
an input for receiving a data word to be stored in the memory;
an encoder operatively coupled with the input, the encoder generating a 16-bit check datum as a function of the data word to be stored and a set of encode data;
a data handler operatively coupled with the memory, the data handler storing the data word in the memory;
a decoder that uses the check datum and the data word in the memory to generate a 16-bit syndrome; and
an error module for determining if the data word in the memory is correct as a function of the syndrome and correction between two and eight non-adjacent bits in the data word.
22. The apparatus as defined by claim 21 wherein the encoder is configured to apply a function to selected bits of the data word, the selected bits being selected as determined by the set of encode data.
23. The apparatus as defined by claim 21 wherein the error module includes:
a comparator for comparing the syndrome to a set of error identity data; and
a corrector for correcting the error if the syndrome matches at least one datum in the set of error identity data, the corrector being configured to correct between one and four bits of data.
24. The apparatus as defined by claim 23 wherein the set of error identity data is generated from the set of encode data, the set of error identity data including a plurality of values assigned to each bit of the data word, the error of a given bit being corrected if the syndrome matches at least one of the plurality of values assigned to the given bit.
25. The apparatus as defined by claim 21 wherein if the error module determines that a given bit of the data word has an error, then it re-reads the given bit to determine if the error has been corrected.
26. The apparatus as defined by claim 25 wherein the data word is stored in a mirror memory location, and if the given bit is determined to have the error after it is re-read, then the error module is configured to retrieve the correct data for the given bit from the mirror memory location and store that corrected data in the given bit.
27. The apparatus as defined by claim 25 wherein the error module is configured to correct other bit errors in the data word if the given bit is corrected.