This article explains the meaning of the "EDAC-sbridge-MC0-HANDLING-MCE-MEMORY-ERROR" error on a cRPD platform.

Users may see the following log messages in dmsg logs on cRPD.

[4306913.247860] EDAC sbridge MC0: HANDLING MCE MEMORY ERROR

When you review the complete log, you can obtain more details about the message as shown below:

[4306873.495978] EDAC MC0: 1 CE memory read error on CPU_SrcID#1_Ha#0_Chan#3_DIMM#0 (channel:3 slot:0 page:0x20531c8 offset:0x40 grain:32 syndrome:0x0 -  area:DRAM err_code:0001:0093 socket:1 ha:0 channel_mask:8 rank:1)
[4306897.573938] {4}[Hardware Error]: Hardware error from APEI Generic Hardware Error Source: 0
[4306897.582485] {4}[Hardware Error]: It has been corrected by h/w and requires no further action
[4306897.591193] {4}[Hardware Error]: event severity: corrected
[4306897.596932] {4}[Hardware Error]:  Error 0, type: corrected
[4306897.602669] {4}[Hardware Error]:  fru_text: DIMM H1
[4306897.607800] {4}[Hardware Error]:   section_type: memory error
[4306897.613805] {4}[Hardware Error]:   physical_address: 0x00000020531c8040
[4306897.620677] {4}[Hardware Error]:   physical_address_mask: 0x00003fffffffffc0
[4306897.628002] {4}[Hardware Error]:   node: 2 card: 3 module: 0 row: 64118 column: 0
[4306897.635844] {4}[Hardware Error]:   error_type: 2, single-bit ECC
[4306897.642150] mce_notify_irq: 15 callbacks suppressed
[4306897.642155] mce: [Hardware Error]: Machine check events logged
[4306912.864685] mce: [Hardware Error]: Machine check events logged
[4306913.247860] EDAC sbridge MC0: HANDLING MCE MEMORY ERROR
[4306913.253371] EDAC sbridge MC0: CPU 10: Machine Check Event: 0 Bank 7: 8c00004000010093
[4306913.261489] EDAC sbridge MC0: TSC 2ee66587451cfc
[4306913.261490] EDAC sbridge MC0: ADDR 2068a48100
[4306913.266463] EDAC sbridge MC0: MISC 142121286
[4306913.271180] EDAC sbridge MC0: PROCESSOR 0:406f1 TIME 1650402847 SOCKET 1 APIC 20

As you can see in above logs, the single-bit ECC error is being reported frequently and is being corrected as well. The logs also indicate the hardware location where the ECC error has occurred:

[4306897.628002] {4}[Hardware Error]: node: 2 card: 3 module: 0 row: 64118 column: 0

If the error occurs only once, you can ignore it. If it is being reported frequently on the same card, reboot the card once because these are transient errors that are usually get rectified after a reboot.

However, if the error continues even after reboot, the module or card where is error is being reported may need replacement.

2022-06-02