System and method for continuous logging of correctable errors without rebooting
In some embodiments, the invention involves a system and method to continuously log correctable errors without rebooting by changing the granularity of the error detection and logging mechanism. A mask register is used to identify which errors are to be logged. Each bit of the mask register may represent a different memory component of the system. Logging of the memory component is determined by the value of the bit in the mask. The masking enables granularity of error logging to the channel and/or dual in-line memory module (DIMM) level. Other embodiments are described and claimed.
1. A system comprising:
a platform comprising a processor and at least one platform component, the at least one platform component being capable of driving correctable errors, the errors capable of being logged; and
a mask register corresponding to the at least one platform component, the mask register to identify a first element of the at least one platform component where error logging is to be disabled, while continuing to enable error logging in a second element of the at least one platform component,
wherein the at least one platform component comprises memory, and
wherein the mask register comprises a bitmap of elements of the memory.
2. The system as recited in claim 1 , wherein the mask register resides on one of a platform motherboard or a controller of the at least one platform component.
3. The system as recited in claim 2 , wherein the controller comprises a memory controller.
4. The system as recited in claim 3 , wherein the memory comprises fully buffered dual in-line memory modules (DIMMs FBD) and the mask register resides in the memory controller, and wherein the mask register comprises a bitmap of the DIMMs.
5. The system as recited in claim 3 , wherein the memory comprises double data rate (DDR) memory and the mask register is coupled to the platform motherboard, and wherein the mask register comprises a bitmap of the DDR memory.
6. The system as recited in claim 1 , wherein the mask register is updated by platform firmware basic input output system (BIOS) runtime handler.
7. The system as recited in claim 6 , wherein an out-of-band controller drives the BIOS to update the mask register to identify elements that are to be disabled from error logging.
8. The system as recited in claim 1 , wherein the error logging of the first element depends on a first threshold to determine a frequency of an error in a predetermined time frame, and a second threshold to determine a count of re-enable of the error logging for the first element.
9. Acomputer implemented method comprising,
initiating runtime logging of correctable errors of a platform component on a computing platform; and
in response to a correctable error event in the computing platform, the error associated with an element of the platform component,
determining whether a threshold for the correctable error event has been reached, and if not, then logging the error, but if so, then disabling error logging for the element by setting a bit in a mask register coupled to the computing platform to indicate that error logging is disabled for the element, the bit being associated with the element, while allowing error logging to continue for other elements of the platform component, and
checking a leaky bucket counter to determine whether the correctable error should be enabled or disabled.
10. The method as recited in claim 9 , wherein the mask register resides on one of a platform motherboard or a controller of the at least one platform component.
11. The method as recited in claim 10 , wherein the platform component comprises memory, and the controller comprises a memory controller, and wherein the memory comprises fully buffered dual in-line memory modules (DIMMs FBD) and the mask register resides in the memory controller, and wherein the mask register comprises a bitmap of the DIMMs.
12. The method as recited in claim 10 , wherein the platform component comprises double data rate (DDR) memory and the mask register is coupled to the platform motherboard, and wherein the mask register comprises a bitmap of the DDR memory.
13. The method as recited in claim 9 , further comprising:
communicating with an out-of-band controller to control the enabling/disabling of correctable errors.
14. A machine readable storage medium having instructions stored therein, that when executed, cause a machine to:
initiate runtime logging of correctable errors on a platform; and
in response to a correctable error event in the platform, the error associated with an element of a platform component,
determine whether a threshold for the correctable error event has been reached, and if not, then log the error, but if so, then disable error logging for the element by setting a bit in a mask register to indicate that error logging is disabled for the element, the bit being associated with the element, while allowing error logging to continue for other elements of the platform component, and
check a leaky bucket counter to determine whether the correctable error should be enabled or disabled.
15. The medium as recited in claim 14 , wherein the mask register resides on one of a platform motherboard or a controller of the at least one platform component.
16. The medium as recited in claim 15 , wherein the platform component comprises memory, and the controller comprises a memory controller, and wherein the memory comprises fully buffered dual in-line memory modules (DIMMs FBD) and the mask register resides in the memory controller, and wherein the mask register comprises a bitmap of the DIMMs.
17. The medium as recited in claim 15 , wherein the platform component comprise double data rate (DDR) memory and the mask register is coupled to the platform motherboard, and wherein the mask register comprises a bitmap of the DDR memory.
18. The medium as recited in claim 14 , further comprising instructions to:
communicate with an out-of-band controller to control the enabling/disabling of correctable errors.