System and method to control memory failure handling on double-data rate dual in-line memory modules
An information handling system includes a processor, a dual in-line memory module (DIMM), and a memory controller coupled to the DIMM. The memory controller provides interrupts to the processor each time a read transaction from the DIMM results in a correctable read error. The processor instantiates a failure predictor to receive the interrupts, accumulate a count of the interrupts, and provide a first error indication when the count exceeds a first error threshold. The failure predictor increments the count each time the predictor receives a particular interrupt and decrements the count in accordance with an error leak rate. The error leak rate has a first value when the DIMM is newer than a first age threshold and has a second value when the DIMM is older than the first age threshold.
1. An information handling system, comprising:
a processor;
a dual in-line memory module (DIMM); and
a memory controller coupled to the DIMM, the memory controller configured to provide interrupts to the processor each time a read transaction from the DIMM results in a correctable read error;
wherein the processor is configured to instantiate a failure predictor, the failure predictor to receive the interrupts, to accumulate a count of the interrupts, and to provide a first error indication when the count exceeds a first error threshold, wherein in accumulating the count, the failure predictor increments the count each time the predictor receives a particular interrupt and decrements the count in accordance with an error leak rate, and wherein the error leak rate has a first value when the DIMM is newer than a first age threshold and has a second value when the DIMM is older than the first age threshold.
2. The information handling system of claim 1 , wherein the DIMM includes a date code that provides a manufacture date of the DIMM.
3. The information handling system of claim 2 , wherein the failure predictor is further to read the date code from the DIMM, and to determine that the DIMM is newer or older than the first age threshold based upon the date code.
4. The information handling system of claim 1 , wherein the memory controller includes an installation date for the DIMM.
5. The information handling system of claim 4 , wherein the failure predictor is further to read the installation date from the memory controller, and to determine that the DIMM is newer or older than the first age threshold based upon the installation date.
6. The information handling system of claim 1 , wherein the failure predictor is further to provide a second error indication when the count exceeds a second error threshold, the second threshold higher than the first error threshold.
7. The information handling system of claim 1 , wherein the error leak rate further maintains the second value when the DIMM is newer than a second age threshold and has a third value when the DIMM is older than the second age threshold, wherein the second age threshold specifies an older age than the first age threshold.
8. The information handling system of claim 1 , wherein the first value of the error leak rate provides a faster leak rate than the second value of the leak rate.
9. The information handling system of claim 1 , wherein the failure predictor is included in an interrupt handler of the information handling system.
10. A method of controlling memory failure handling, the method comprising:
providing, by a memory controller of an information handling system, interrupts to a processor of the information handling system each time a read transaction between the memory controller and a dual in-line memory module (DIMM) of the information handling system results in a correctable read error;
receiving, by a failure predictor instantiated by the processor, the interrupts;
accumulating, by the failure predictor, a count of the interrupts; and
providing, by the failure predictor, a first error indication when the count exceeds a first error threshold;
wherein the accumulating further includes:
incrementing the count each time the predictor receives a particular interrupt; and
decrementing the count in accordance with an error leak rate, wherein the error leak rate has a first value when the DIMM is newer than a first age threshold and has a second value when the DIMM is older than the first age threshold.
11. The method of claim 10 , wherein the DIMM includes a date code that provides a manufacture date of the DIMM.
12. The method of claim 11 , further comprising:
reading, by the failure predictor, the date code from the DIMM; and
determining, by the failure predictor, that the DIMM is newer or older than the first age threshold based upon the date code.
13. The method of claim 10 , wherein the memory controller includes an installation date for the DIMM.
14. The method of claim 13 , further comprising:
reading, by the failure predictor, the installation date from the memory controller; and
determining, by the failure predictor, that the DIMM is newer or older than the first age threshold based upon the installation date.
15. The method of claim 10 , further comprising:
providing, by the failure predictor, a second error indication when the count exceeds a second error threshold, the second threshold higher than the first error threshold.
16. The method of claim 10 , wherein the error leak rate further maintains the second value when the DIMM is newer than a second age threshold and has a third value when the DIMM is older than the second age threshold, wherein the second age threshold specifies an older age than the first age threshold.
17. The method of claim 10 , wherein the first value of the error leak rate provides a faster leak rate than the second value of the leak rate.
18. The method of claim 10 , wherein the failure predictor is included in an interrupt handler of the information handling system.
19. An information handling system, comprising:
a processor;
a hardware management engine;
a dual in-line memory module (DIMM); and
a memory controller coupled to the DIMM, the memory controller including a count of a number of times a read transaction from the DIMM resulted in a correctable read error;
wherein the hardware management engine is configured to instantiate a failure predictor to retrieve a first number of correctable read errors from the memory controller, to determine a difference between the first number of correctable read errors and a second number of correctable read errors retrieved by the failure predictor, to adjust the difference in accordance with an error leak rate, and to provide an interrupt to the processor when the adjusted difference exceeds an error threshold, wherein the error leak rate has a first value when the DIMM is newer than a first age threshold and has a second value when the DIMM is older than the first age threshold.
20. The information handling system of claim 19 , wherein:
the memory controller includes a read counter that provides a number of read operations the memory controller has performed on the DIMM; and
the failure predictor is further to read the read counter from the memory controller.