Method and apparatus for transferring data between a host processor and a data storage device

• US 4,745,604 A
• Filed: 10/20/1986
• Issued: 05/17/1988
• Est. Priority Date: 10/20/1986
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Yu, Wellington C., Wang, David T., Patel, Arvind M.
• Primary Examiner(s): ATKINSON, CHARLES
• Application Number:
  US06/921,022
• Time in Patent Office: 575 Days
• Field of Search: 371/37, 371/38, 371/39, 371/40, 364/200, 364/900
• US Class Current: 714/753
• CPC Class Codes: G06F 11/10 Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's
• CPC Class Codes: G06F 3/0601 Interfaces specially adapted for storage systems
• CPC Class Codes: G06F 3/0673 Single storage device
• CPC Class Codes: H03M 13/03 Error detection or forward error correction by redundancy in data representation, i.e. code words containing more digits than the source words

Abstract

In a data processing system in which two-level error correction is performed on variable length data being transferred between the host processor and the data storage device, the logical length of the data being transferred is computed during a fixed time gap with computation continuing after termination of the fixed time gap and commencement of the data transfer. The computation required for the logical length of the data field to accommodate two-level ECC is accomplished by first comparing the actual field length with a value predetermined by the subblock length of the two-level ECC. If the actual length is greater than the predetermined value, then a value equal to the subblock length plus first level ECC bytes is loaded into a counter which begins decrementing at the termination of the fixed time gap so as to synhronize by the byte-by-byte transfer of the data. As the data is being transferred the computation continues. When the computation has been completed, a value equal to the difference between the computed logical length and the subblock length plus first level ECC bytes is loaded into other counters which begin decrementing when the first counter reaches zero. In this manner transfer of the data is not interrupted. This permits two-level ECC to be incorporated into prior data processing systems which utilize a conventional track format with predetermined fixed time gaps, even though such fixed time gaps would otherwise be of insufficient duration to permit computation of the logical length required for two-level ECC.

Claims

1. A method for transferring data having variable length fields which are separated by fixed time gaps between a host processor and a data storage device wherein error correction bytes are encoded into or decoded from the data during transfer, the method comprising the steps of:

determining the actual length of the field to be transferred;

computing the number of error correction bytes for the field being transferred;

initiating transfer of the field at the termination of the fixed time gap and before said computation has been completed if the determined actual field length is greater than a predetermined value; and

completing transfer of the field when said computation has been completed.

2. The method according to claim 1 further comprising the steps of computing the number of padding bytes required for the field and computing the logical length of the field from the determined actual length, the computed error correction bytes and the computed padding bytes.

3. The method according to claim 1 wherein the step of initiating transfer of the field further comprises decrementing a count equal to a fixed value related to the predetermined value, and wherein the step of completing transfer of the field further comprises decrementing a count equal to the difference between the logical length of the field and the fixed value when the decremented count of the fixed value equals zero.

4. The method according to claim 1 wherein the error correction bytes are encoded and decoded at a subblock level and a field level, and wherein the step of computing the number of error correction bytes further comprises determining the number of subblocks within the field.

5. A method for transferring variable length data fields between a host processor and a disk file wherein two-level error correction is encoded into or decoded from the fields during transfer of the fields, the method comprising the steps of:

determining the actual length of the field to be transferred;

computing the number of subblocks within the field;

computing the number of error correction bytes required at the subblock and field levels;

initiating transfer of the field before computation of the number of required correction bytes is completed if the determined actual field length is greater than a predetermined value; and

completing transfer of the field when the error correction byte computation has been completed.

6. The method according to claim 5 wherein the step of initiating transfer of the field further comprises transferring initial bytes in the field while simultaneously counting down from a fixed value related to the predetermined value, and wherein the step of completing the transfer of the data field further comprises transferring remaining bytes in the field when the counting reaches zero.

7. The method according to claim 5 further comprising the steps of computing the number of padding bytes required for the field and summing the computed number of padding bytes, the number of error correction bytes and the actual length.

8. In an apparatus for the transfer of information of variable field length between a host processor and a data recording disk file wherein multiple-level error correction encoding and decoding of the fields into subblocks during transfer requires that the logical length of the fields be calculated during fixed time gaps between successive fields, an improvement comprising:

means coupled to the information transfer apparatus for reading the length of the field to be transferred at the end of the fixed time gap;

means for computing the number of subblocks in the field whose length has been read during the fixed time gap and for computing the logical length of the field required for error correction;

means for comparing the length of the field read with a predetermined value;

a first counter coupled to the computing means for receiving a fixed value related to the predetermined value from the computing means and for initiating transfer of the field before computation of the logical length has been completed by the computing means; and

a second counter coupled to the computing means for completing the transfer of the field when the first counting means has terminated the count of the fixed value.

9. The improvement according to claim 8 wherein at least some of the tracks in the disk file have defects preventing the writing or reading of data, the improvement further comprising:

memory storage means coupled to the computing means for storing values of the size and location of defects on a track onto which the subblocks of the field are to be written to or read from; and

wherein the computing means further comprises means responsive to the stored values for splitting the subblocks into segments, whereby the subblock segments may be written to or read from the tracks across the defects in the tracks, and means for calculating the logical length required for error correction of each subblock split into segments.

10. The improvement according to claim 8 wherein the information transfer apparatus is of the type wherein each field is padded to a discrete number of cells and wherein the computing means further comprises means for computing the logical length of the field required for both error correction and cell padding.

11. The improvement according to claim 8 wherein the second counter further comprises means for receiving from the computing means a value equal to the difference between the computed logical length and the fixed value.