Margining decoding utilizing soft-inputs

1. A method for processing a portion of a storage device, comprising:

obtaining known data associated with the portion;

reading back from the portion to produce a read-back waveform;

using a detector to decode the read-back waveform, including by producing statistical information which includes (1) a set of one or more estimated values associated with the portion and (2) a set of one or more reliabilities, wherein each reliability in the set of reliabilities indicates a reliability with which a corresponding estimated value in the set of estimated values is generated;

determining which of the set of estimated values are incorrect based at least in part on the set of estimated values and the known data; and

counting the number of incorrect estimated values which have a corresponding reliability which is greater than a reliability threshold, including by:

determining a number of incorrect log-likelihood ratios (LLRs) that have a magnitude greater than a first threshold; and

comparing the number of incorrect LLRs that have a magnitude greater than the first threshold against a second threshold.

2. A method as recited in claim 1 , wherein obtaining includes writing known data to the portion.

3. A method as recited in claim 1 , wherein the storage device is a hard disk drive and the portion is a sector.

4. A method as recited in claim 1 further comprising:

determining whether the portion is defective, including by:

comparing the count of incorrect estimated values which have a reliability greater than the reliability threshold to a count threshold;

in the event the count of incorrect estimated values which have a reliability greater than the reliability threshold is less than or equal to the count threshold, determining that the portion is not defective; and

in the event the count of incorrect estimated values which have a reliability greater than the reliability threshold is greater than the count threshold, determining that the portion is defective.

5. A method as recited in claim 4 , wherein:

determining whether the portion is defective is performed during manufacturing testing; and

the method further includes: in the event the count of incorrect estimated values which have a reliability greater than the reliability threshold is greater than the count threshold, adding the defective portion to list of defects detected during manufacturing testing.

6. A method as recited in claim 4 , wherein:

determining whether the portion is defective is performed during field operation; and

the method further includes: in the event the count of incorrect estimated values which have a reliability greater than the reliability threshold is greater than the count threshold, moving data to a different location.

7. A method as recited in claim 4 further comprising determining the count threshold based at least in part on an error correction capability of a code, including by generating a higher count threshold for a first code which can correct more errors than for a second code which can correct fewer errors.

8. A system for processing a portion of a storage device, comprising:

a processor configured to:

obtain known data associated with the portion;

read back from the portion to produce a read-back waveform;

decode the read-back waveform, including by producing statistical information which includes (1) a set of one or more estimated values associated with the portion and (2) a set of one or more reliabilities, wherein each reliability in the set of reliabilities indicates a reliability with which a corresponding estimated value in the set of estimated values is generated;

determine which of the set of estimated values are incorrect based at least in part on the set of estimated values and the known data; and

count the number of incorrect estimated values which have a corresponding reliability which is greater than a reliability threshold, including by:

determining a number of incorrect log-likelihood ratios (LLRs) that have a magnitude greater than a first threshold; and

comparing the number of incorrect LLRs that have a magnitude greater than the first threshold against a second threshold; and

a communication interface coupled to the processor.

9. A system as recited in claim 8 , wherein the processor is further configured to: determine whether the portion is defective, including by:

comparing the count of incorrect estimated values which have a reliability greater than the reliability threshold to a count threshold;

in the event the count of incorrect estimated values which have a reliability greater than the reliability threshold is less than or equal to the count threshold, determining that the portion is not defective; and

in the event the count of incorrect estimated values which have a reliability greater than the reliability threshold is greater than the count threshold, determining that the portion is defective.

10. A system as recited in claim 9 , wherein:

determining whether the portion is defective is performed during manufacturing testing; and

the processor is further configured to: in the event the count of incorrect estimated values which have a reliability greater than the reliability threshold is greater than the count threshold, add the defective portion to list of defects detected during manufacturing testing.

11. A system as recited in claim 9 , wherein:

determining whether the portion is defective is performed during field operation; and

the processor is further configured to: in the event the count of incorrect estimated values which have a reliability greater than the reliability threshold is greater than the count threshold, move data to a different location.

12. A system as recited in claim 9 , wherein the processor is further configured to: determine the count threshold based at least in part on an error correction capability of a code, including by generating a higher count threshold for a first code which can correct more errors than for a second code which can correct fewer errors.

13. A computer program product for processing a portion of a storage device, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

obtaining known data associated with the portion;

reading back from the portion to produce a read-back waveform;

decoding the read-back waveform, including by producing statistical information which includes (1) a set of one or more estimated values associated with the portion and (2) a set of one or more reliabilities, wherein each reliability in the set of reliabilities indicates a reliability with which a corresponding estimated value in the set of estimated values is generated;

determining which of the set of estimated values are incorrect based at least in part on the set of estimated values and the known data; and

counting the number of incorrect estimated values which have a corresponding reliability which is greater than a reliability threshold, including by:

determining a number of incorrect log-likelihood ratios (LLRs) that have a magnitude greater than a first threshold; and

comparing the number of incorrect LLRs that have a magnitude greater than the first threshold against a second threshold.

14. A computer program product as recited in claim 13 further comprising computer instructions for: determining whether the portion is defective, including computer instructions for:

comparing the count of incorrect estimated values which have a reliability greater than the reliability threshold to a count threshold;

in the event the count of incorrect estimated values which have a reliability greater than the reliability threshold is less than or equal to the count threshold, determining that the portion is not defective; and

in the event the count of incorrect estimated values which have a reliability greater than the reliability threshold is greater than the count threshold, determining that the portion is defective.

15. A computer program product as recited in claim 14 , wherein:

determining whether the portion is defective is performed during manufacturing testing; and

the computer program product further includes computer instructions for: in the event the count of incorrect estimated values which have a reliability greater than the reliability threshold is greater than the count threshold, adding the defective portion to list of defects detected during manufacturing testing.

16. A computer program product as recited in claim 14 , wherein:

determining whether the portion is defective is performed during field operation; and

the computer program product further includes computer instructions for: in the event the count of incorrect estimated values which have a reliability greater than the reliability threshold is greater than the count threshold, moving data to a different location.

17. A computer program product as recited in claim 14 further comprising computer instructions for: determining the count threshold based at least in part on an error correction capability of a code, including by generating a higher count threshold for a first code which can correct more errors than for a second code which can correct fewer errors.