Circuit and method for rapid checking of error correction codes using cyclic redundancy check

Abstract

A circuit and method use a cyclic redundancy check (CRC) unit to check any errors detected by an error correction code (ECC) unit when reading a sector of bytes from a disk in a disk drive. The CRC unit reads the sector of bytes from the disk simultaneously with the ECC unit. The CRC unit begins generating a residue to detect errors in the sector of bytes at approximately the same time the ECC unit begins generating a residue to detect errors in the sector. The CRC unit does not wait for the ECC unit to finish correcting the sector and transfer the ECC error corrections into a buffer memory before generating a CRC residue. The CRC unit monitors both the sector of bytes read from the disk platter and any errors detected by the ECC unit in the data and CRC bytes, without reading the contents of the buffer unit. The CRC unit uses the error locations and error values found by the ECC unit to determine whether the ECC-corrected bytes stored in a buffer unit were properly corrected. The CRC unit compares any errors detected by the ECC unit with any errors detected by the CRC unit. If the errors detected by the CRC unit do not match the errors detected by the ECC unit, then the CRC unit causes the disk drive to discard the sector of bytes and attempt to reread the same sector from the disk. The CRC error check is completed before any data is transferred to the host computer. The CRC unit must approve of the data before the disk drive controller sends the data to the host. The ECC unit may begin to process another sector of data while the CRC unit is comparing any errors detected by the ECC unit with any errors detected by the CRC unit.

Dependencies are inferred from the claim text.