Disk controller architecture to allow on-the-fly error correction and write disruption detection
View Patent ↗Error correction in a disk drive is performed by error correction circuitry which accepts data read from a data storage medium. The error correction circuitry performs both block error correction in a first data domain and sector error correction in a second data domain. A sector FIFO buffer is used to facilitate the error correction in real time, or “on-the-fly.” The sector FIFO buffer also enables conversion of the corrected data to the first data domain. The error correction circuitry also generates an ECC block comprising a plurality of sectors and writes the ECC block. The circuitry generates a tag prior to writing the ECC block and adds the tag to each of a plurality of sectors. During a read operation, the circuitry detects a write disruption when the tags for all of the plurality of sectors in the ECC block are not identical.
1. A method for determining disruption during the write operation of an ECC block comprising:
generating a tag at a start of a write operation of the ECC block;
adding the tag to each of a plurality of sectors which comprise the ECC block during the write operation;
comparing the tags for all of the plurality of sectors in the ECC block at any time after the write operation is complete; and
determining a write disruption occurred if the tags for all of the plurality of sectors in the ECC block are not identical.
2. The method of claim 1 , further comprising generating a notification if a write disruption is detected.
3. The method of claim 1 , wherein the tag is generated using a random number generator.
4. The method of claim 1 , further comprising logging write disruptions indicated by non- matching tags in memory.
5. The method of claim 1 , further comprising determining whether a write error occured based on whether the tags for all of the plurality of sectors in the ECC block contain identical values.
6. The method of claim 1 , further comprising determining that no write error occurred when the comparison step indicates that all of the tags in the plurality of sectors in the ECC block contain identical values.
7. The method of claim 1 , further comprising determining that a write error occurred when the comparison step indicates that any of the tags in the plurality of sectors in the ECC block contain non-identical values.
8. A disk drive comprising:
a read/write head;
a data storage medium; and
control circuitry which transfers data from the data storage medium using the read/write head and performs error correction on the data to generate an ECC block comprising a plurality of sectors and writes the ECC block, wherein the control circuitry generates a tag prior to writing the ECC block and adds the tag to each of the plurality of sectors, wherein the control circuitry detects a write disruption when the tags for all of the plurality of sectors in the ECC block are not identical after writing.
9. The disk drive of claim 8 , wherein the tags generated by the control circuit are modified at the beginning of any write operation.
10. The disk drive of claim 9 , wherein the tags are randomly generated.
11. The disk drive of claim 8 , wherein the control circuit generates an error notice if a write disruption occurs.