Multilevel memory system with copied error detection bits
In described examples, a memory system is accessed by reading a data line and error detection bits for the data line from a first memory. The data line and the error detection bits from the first memory are decoded to determine if an error is present in the data line from the first memory. A copy of the data line and the error detection bits are stored in a second memory. The copy of the data line and error detection bits are read from the second memory. The copy of the data line and error detection bits are decoded to determine if an error is present in the copy of the data line from the second memory.
1. A method of operating a memory system, the method comprising:
reading a data line and error detection bits for the data line from a first memory;
decoding the data line and the error detection bits from the first memory to determine if an error is present in the data line;
storing a copy of the data line and the error detection bits read from the first memory in a second memory;
reading the copy of the data line and error detection bits from the second memory; and
decoding the copy of the data line and error detection bits to determine if an error is present in the copy of the data line read from the second memory.
2. The method of claim 1 , wherein decoding the data line and error detection bits from the first memory is done by a first decoder and by a second decoder; and further comprising detecting a decoder error by comparing results from the first decoder and the second decoder.
3. The method of claim 2 , wherein decoding the copy of the data line and error detection bits from the second memory is done by a third decoder and by a fourth decoder; and further comprising detecting a decoder error by comparing results from the third decoder and the fourth decoder.
4. The method of claim 2 , wherein decoding the data line and error detection bits generates a syndrome.
5. The method of claim 1 , further comprising:
receiving the data line and calculating the error detection bits; and
storing the data line and error detection bits in the first memory.
6. The method of claim 1 , wherein if a one-bit error is detected in the data line from the first memory, further comprising correcting the one-bit error using the error detection bits.
7. The method of claim 1 , wherein if an error is detected in the copy of the data line, further comprising re-reading the data line and error detection bits from the first memory and replacing the copy of the data line and error detection bits in the second memory with another copy of the data line and error detection bits.
8. The method of claim 1 , wherein the first memory is in a lower level of a multilevel memory system and the second memory is in a higher level of the multilevel memory system.
9. The method of claim 8 , wherein the first memory is a cache memory and the second memory is a cache memory.
10. The method of claim 1 , wherein the first memory is a main memory and the second memory is a cache memory.
11. A system comprising:
a first memory configured to store multiple lines of multibit (N) data and respective multiple (M) error detection bits;
an error detection and correction code (ECC) module coupled to the first memory, the ECC module having an output to provide corrected N-bit data and respective M-bit error detection bits for a line of N-bit data accessed from the first memory;
a second memory configured to store multiple lines of N-bit data and respective multiple M-bit error detection bits, wherein the second memory is coupled to the ECC module output and configured to store a copy of the corrected N-bit data and M-bit error bits; and
an error detection (ED) module coupled to the second memory, the ED module having an output to indicate an error in a line of N-bit data read from the second memory.
12. The system of claim 11 , wherein the first memory is a main memory and the second memory is a cache memory.
13. The system of claim 11 , wherein the ECC module is a first ECC module, further comprising:
a second ECC module coupled to the first memory, wherein the first ECC module as a first error syndrome output and the second ECC module has a second error syndrome output; and
a diagnostic module with a first input coupled to first error syndrome output and a second input coupled to the second error syndrome output, the diagnostic module having an output to indicate when a first error syndrome value is different from a second error syndrome value.
14. The system of claim 13 , wherein the ED module is a first ED module, further comprising:
a second ED module coupled to the second memory, wherein the first ED module has a third error syndrome output and the second ED module has a fourth error syndrome output; and
wherein the diagnostic module has a third input coupled to the third error syndrome output and a fourth input coupled to the fourth error syndrome output.
15. The system of claim 11 , wherein the first memory has at least two banks configured to each store N-bit data and respective M-bit error detection bits and the ECC module has a respective at least two portions that operate in parallel.