Iterative decoding technique for correcting DRAM device failures
Provided is a memory system comprising a plurality of memory components; and a controller in communication with the plurality of memory components and configured to perform error correction code (ECC) decoding on a received word read from the plurality of memory components. The ECC decoding is configured to (i) detect one or more random errors in a portion of the received word, the portion corresponding to one of the components within the plurality, and (ii) correct the detected random errors; and when the correcting of the detected random errors fails, iteratively marking symbols in the remaining portions of the received word as erasures.
1. An error correction code (ECC) decoder configured to receive data from a memory, comprising:
a syndrome generator configured to receive a codeword from the memory;
an error-location polynomial generator coupled to an output of the syndrome generator; an error value generator; and
an error location generator;
wherein the error value generator and the error location generator are electrically coupled to an output of the error-location polynomial generator;
wherein the error value generator and the error location generator (i) facilitate performing a single decoding attempt to detect random error symbols in a first portion of a word received from one memory component of a plurality of memory components of the memory and (ii) correct the detected random error; and
wherein when the correcting of the detected random errors fails, iteratively performing decoding attempts with each attempt marking symbols in the remaining portions of the received word that correspond respectively to remaining memory components of the plurality of memory components, as erasures, and using counter logic to obtain erasure location information corresponding to the erasures.
2. The ECC decoder of claim 1 , wherein the ECC decoder is formed within a memory controller, the memory controller being coupled, at least indirectly, to the memory.
3. The ECC decoder of claim 1 , wherein the ECC decoder is based upon Reed Solomon (RS) codes.
4. The ECC decoder of claim 3 , wherein the memory includes dynamic random access memory (DRAM) devices.
5. The ECC decoder of claim 4 , wherein the memory is constructed in accordance compute express link (CXL) principles.
6. The ECC decoder of claim 1 , wherein the syndrome generator outputs erasure values to the error-location polynomial generator.
7. The ECC decoder of claim 6 , wherein the error-location polynomial generator is based upon one from the group including a Berlekamp algorithm and a Euclid algorithm.
8. The ECC decoder of claim 7 , wherein the error value generator operate in accordance with Forney algorithm principles.
9. The ECC decoder of claim 8 , wherein the search is performed in accordance with Chien search algorithm principles.