Dynamic data transfer bandwidth control

• US 6,079,046 A
• Filed: 04/09/1998
• Issued: 06/20/2000
• Est. Priority Date: 04/09/1997
• Status: Expired due to Term

• Current Assignee: Mitsubishi Electric Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Yamada, Kohji, Itagaki, Hiroshi, Tamura, Tetsuya, Demura, Masayuki
• Primary Examiner(s): MOISE, EMMANUEL LIONEL
• Application Number:
  US09/057,998
• Time in Patent Office: 803 Days
• Field of Search: 714/781, 714/785, 714/774, 375/219, 375/222, 370/209, 370/342, 370/433
• US Class Current: 714/781
• CPC Class Codes: G11B 20/10527 Audio or video recording; Data buffering arrangements G11B20/12 - G11B20/18 take precedence
• CPC Class Codes: G11B 20/1813 by adding special bits or symbols to the coded information G11B20/1809 takes precedence

Abstract

It is one object of the present invention to perform efficient data transfer processing wherein, during decoding using a combination of product coding and erasure correction, real-time correction can be performed. To achieve the above object, provided is a method whereby a transfer mode is dynamically altered, in accordance with the state of a decoder, and in additional buffer, consisting of two banks, is added, so that the speed for the data transferring relative to the decoder is increased.

Claims

1. In a system, wherein a first buffer, a second buffer, a decoder and an MPU are connected to each other by a bus, a method of transferring data from a storage medium to a host, the data being initially read from the storage medium and stored in the first buffer, the data being encoded in a product code for error detection and correction, the method comprising the steps of:

(a) transferring a subset of the data in the product code from the first buffer to said second buffer;

(b) transferring a subset of the data from the second buffer to the decoder;

(c) while the subset of data is being received by the decoder, determining a number of errors in the data revealed by the product code;

(d) if the number of errors exceeds a threshold, increasing a portion of the system bandwidth used for error correction.

2. The method according to claim 1, further comprising the step of inserting a wait before a transfer of data from said first buffer to said second buffer is resumed if the number of errors exceeds a selected value.

3. The method according to claim 1, wherein step (a) transfers subsets of the data in a page mode.

4. The method according to claim 1, wherein the number of errors is found by said decoder using an error location polynomial and an error evaluation polynomial provided by a Euclidian algorithm.

5. The method according to claim 1, wherein said transfer step (b) further comprises selecting one of two or more banks comprising the second buffer to hold the data, and step (c) further includes the step of detecting when one bank is filled with data and the counting of errors in the bank is completed in a period during which data is transferred to one of more remaining banks.

6. The method according to claim 1, wherein the subsets of the data are columns and rows and when row data in said first buffer is transferred to said second buffer, a wait between said transfers of data is set by said MPU.

7. The method according to claim 1, wherein the transfer of subsets of data from said first buffer to said second buffer is performed in a burst mode.

8. The method according to claim 1, further comprising the steps of:

calculating syndromes for error correction;

calculating an error location polynomial and an error evaluation polynomial provided by a Euclidian algorithm; and

calculating error locations and values using a Chien search.

9. The method according to claim 8, wherein the step of calculating syndromes is performed while data is being transferred said second buffer to said decoder, so that said transfer of data from said second buffer and said Chien search are pipelined.

10. A system for reading encoded data from a storage medium, comprising:

first and second buffers which are interconnected to allow data to pass from the first buffer to the second buffer;

a decoder which counts errors in the data and is connected to the first and second buffers by a bus and is also directly connected to the second buffer;

an MPU connected to first and second buffers, and the decoder by the bus;

a

means for transferring columns of data in said first buffer to said second buffer;

means for transferring rows of data in said first buffer to said second buffer; and

means for increasing a bandwidth allocation used for error correction when the number of errors exceeds a specified count.

11. The system according to claim 10, further comprising, means for inserting a wait before a transfer of data from said first buffer to said second buffer if the errors exceed a selected threshold.

12. The system according to claim 10, wherein said transfer means for transferring columns of data uses a page mode.

13. The system according to claim 10, further comprising means for calculating the error location polynomial and the error evaluation polynomial by a Euclidian algorithm.

14. The system according to claim 10, wherein the second buffer further comprises two or more banks, and the system further comprises means for beginning calculation of error numbers and error locations for data when one bank is filled with data, and for completing said calculation in a period during which data is transferred to the remaining banks.

15. The system according to claim 10, wherein, when said row data in said first buffer is transferred to said second buffer, a wait is set between said transfers of data.

16. The system according to claim 10, the means for transferring columns uses a burst mode.

17. The system according to claim 10, wherein the decoder includes means for calculating syndromes, means for calculating an error location polynomial and means for calculating an error evaluation polynomial provided by an Euclidian algorithm, and means for calculating error locations and values using a Chien search.

18. The system according to claim 17, wherein said syndrome calculation is performed at the same time as is a transfer of data from said second buffer to said decoder, so that data for a plurality of rows or a plurality of columns are present in said second buffer, and so that said transfer of data from said second buffer and said Chien search are pipelined.

19. A DVD player system for playing data recorded on a DVD, said system comprising:

a motor for rotating said DVD;

an optical pick-up for optically reading said data recorded on said DVD;

drive means for driving said motor and said optical pick-up;

decoding means for correcting errors and transferring decoded data to a host; and

DVD controller means for transferring optical signal from said pick-up to said decoding means, and sending a servo operation command to said drive means based on control information within said data or control signal from said decoding means;

said decoding means, wherein a first buffer, a second buffer, a decoder and an MPU are connected to each other through a bus, further comprising

(a) means for transferring column data in said first buffer to said second buffer,

(b) means for starting a transfer of data from said second buffer to said decoder upon the detection of the number of errors,

(c) means for performing error correction for column data in said first buffer after next data transfer from said first buffer to said second buffer has started,

(d) means for transferring row data in said first buffer to said second buffer, and

(e) means for altering a bandwidth of said host to a lower level and for dynamically altering a bandwidth between said MPU and said first buffer to an upper level when the number of errors in one row on said first buffer exceeds a specified number.

20. The system according to claim 19, further comprising, if many errors exist in column in said first buffer, means (f) for inserting a wait before a transfer of data from said first buffer to said second buffer is resumed.

21. The system according to claim 19, wherein said transfer means (a) is means for transferring data in a page mode.

22. The system according to claim 19, wherein in said means (b) the number of errors is calculated by said decoder using the error location polynomial and the error evaluation polynomial provided by the Euclidian algorithm.

23. The system according to claim 19, wherein said transfer means (d) includes means for dividing said second buffer into two or more banks, for beginning calculation of errors and error locations for data when one bank is filled with data, and for completing said calculation in a period during which data is transferred to the remaining banks.

24. The system according to claim 19, wherein, when said row data in said first buffer is transferred to said second buffer, in said means (d) a wait is set between said transfers of data from said MPU.

25. The system according to claim 24, wherein in said means (d) said wait for data transfer is set to 0, and a transfer of data from said first buffer to said second buffer is performed in a burst mode.

26. The system according to claim 19, wherein processing performed by said decoder includes three stages:

syndrome calculation, calculation of the error location polynomial and the error evaluation polynomial provided by the euclidian algorithm, and calculation of error locations and values by using a Chien search.

27. The system according to claim 26, wherein said syndrome calculation is performed at the same time as is a transfer of data from said second buffer to said decoder, so that data for a plurality of rows or a plurality of columns are present in said second buffer, and so that said transfer of data from said second buffer and said Chien search are pipelined.