Error correction decoder

1. An error correction decoder, comprising:

a mapper configured to generate, based on a first set of read values corresponding to a first codeword, a first set of log likelihood ratio (LLR) values;

a first buffer, coupled to the mapper, configured to store the first set of LLR values received from the mapper;

a node processor, coupled to the first buffer, configured to perform a first error correction decoding operation using the first set of LLR values received from the first buffer; and

a second buffer coupled to the node processor and the mapper,

wherein a first iteration of the first error correction decoding operation comprises:

holding values of one or more variable nodes during updating values of one or more check nodes, and

performing a syndrome check using a parity check matrix and sign bits of the first set of LLR values stored in the first buffer, and

wherein the node processor is further configured to:

perform iterations subsequent to the first iteration upon a determination that the syndrome check has failed in the first iteration of the first error correction decoding operation, and

store the values of the one or more variable nodes that are updated in the iterations subsequent to the first iteration in the second buffer.

2. The error correction decoder of claim 1 , further comprising:

a decoding controller, coupled to the node processor, to the mapper, and to the first buffer and the second buffer, configured to store the sign bits in the second buffer upon a determination that the syndrome check has passed in the first iteration of the first error correction decoding operation.

3. The error correction decoder of claim 2 , wherein the decoding controller is further configured to output the sign bits stored in the second buffer as a decoded codeword.

4. The error correction decoder of claim 3 , wherein the mapper is further configured, prior to the sign bits stored in the second buffer being output, to:

generate a second set of LLR values based on a second set of read values corresponding to second codeword; and

store the second set of LLR values in the first buffer.

5. The error correction decoder of claim 4 , wherein the node processor is further configured to perform a second error correction decoding operation using the second set of LLR values stored in the first buffer, and the decoding controller is further configured, prior to a first iteration of the second error correction decoding operation being performed, to output the sign bits stored in the second buffer.

6. The error correction decoder of claim 1 ,

wherein the iterations subsequent to the first iteration are performed upon a determination that a maximum number of iterations has not been exceeded.

7. The error correction decoder of claim 6 , wherein the node processor is further configured to perform a syndrome check using the parity check matrix and the values of the variable nodes stored in the second buffer in the iterations subsequent to the first iteration.

8. The error correction decoder of claim 7 , wherein a decoding controller is configured to output the values of the variable nodes stored in the second buffer as a decoded codeword upon a determination that the syndrome check based on the values of the variable nodes stored in the second buffer has passed.

9. The error correction decoder of claim 8 , wherein the mapper is further configured, prior to the values of the variable nodes stored in the second buffer being output, to:

generate a second set of LLR values based on a second set of read values corresponding to a second codeword, and

store the second set of LLR values in the first buffer.

10. The error correction decoder of claim 1 , further comprising:

a decoding controller configured to output the sign bits as a decoded codeword upon a determination that the syndrome check has passed in the first iteration of the first error correction decoding operation.

11. The error correction decoder of claim 10 ,

wherein the mapper is further configured to:

generate a second set of LLR values based on a second set of read values corresponding to a second codeword, and

store the second set of LLR values in the second buffer.

12. The error correction decoder of claim 11 , wherein the mapper is configured, prior to the sign bits stored in the first buffer being output, to generate the second set of UR values and store the second set of LLR values in the second buffer.

13. The error correction decoder of claim 11 , wherein the node processor is further configured to perform a second error correction decoding operation using the second set of LLR values stored in the second buffer.

14. The error correction decoder of claim 13 , wherein the decoding controller is further configured, prior to a first iteration of the second error correction decoding operation being performed, to output the sign bits stored in the first buffer.

15. The error correction decoder of claim 1 , wherein the first error correction decoding operation comprises a vertically shuffled scheduling operation.

16. A method for error correction decoding, comprising:

generating, based on a first set of read values corresponding to a first codeword, a first set of log likelihood ratio (LLR) values;

storing the first set of LLR values in a first buffer;

performing a first error correction decoding operation using the first set of LLR values,

wherein a first iteration of the first error correction decoding operation comprises:

holding values of one or more variable nodes during updating values of one or more check nodes, and

performing a syndrome check using a parity check matrix and sign bits of the first set of LLR values stored in the first buffer;

performing iterations subsequent to the first iteration upon a determination that the syndrome check has failed in the first iteration of the first error correction decoding operation; and

storing the values of the one or more variable nodes that are updated in the iterations subsequent to the first iteration in a second buffer.

17. The method of claim 16 , further comprising:

storing the sign bits in the second buffer upon a determination that the syndrome check has passed in the first iteration of the first error correction decoding operation.

18. The method of claim 16 ,

wherein the iterations subsequent to the first iteration are performed upon a determination that a maximum number of iterations has not been exceeded.

19. The method of claim 16 , further comprising: outputting the sign bits as a decoded codeword upon a determination that the syndrome check has passed in the first iteration of the first error correction decoding operation.

20. The method of claim 16 , wherein the first error correction decoding operation comprises a vertically shuffled scheduling operation.