120 lines · plain
1PCIe Device AER statistics2--------------------------3 4These attributes show up under all the devices that are AER capable. These5statistical counters indicate the errors "as seen/reported by the device".6Note that this may mean that if an endpoint is causing problems, the AER7counters may increment at its link partner (e.g. root port) because the8errors may be "seen" / reported by the link partner and not the9problematic endpoint itself (which may report all counters as 0 as it never10saw any problems).11 12What: /sys/bus/pci/devices/<dev>/aer_dev_correctable13Date: July 201814KernelVersion: 4.19.015Contact: linux-pci@vger.kernel.org, rajatja@google.com16Description: List of correctable errors seen and reported by this17 PCI device using ERR_COR. Note that since multiple errors may18 be reported using a single ERR_COR message, thus19 TOTAL_ERR_COR at the end of the file may not match the actual20 total of all the errors in the file. Sample output::21 22 localhost /sys/devices/pci0000:00/0000:00:1c.0 # cat aer_dev_correctable23 Receiver Error 224 Bad TLP 025 Bad DLLP 026 RELAY_NUM Rollover 027 Replay Timer Timeout 028 Advisory Non-Fatal 029 Corrected Internal Error 030 Header Log Overflow 031 TOTAL_ERR_COR 232 33What: /sys/bus/pci/devices/<dev>/aer_dev_fatal34Date: July 201835KernelVersion: 4.19.036Contact: linux-pci@vger.kernel.org, rajatja@google.com37Description: List of uncorrectable fatal errors seen and reported by this38 PCI device using ERR_FATAL. Note that since multiple errors may39 be reported using a single ERR_FATAL message, thus40 TOTAL_ERR_FATAL at the end of the file may not match the actual41 total of all the errors in the file. Sample output::42 43 localhost /sys/devices/pci0000:00/0000:00:1c.0 # cat aer_dev_fatal44 Undefined 045 Data Link Protocol 046 Surprise Down Error 047 Poisoned TLP 048 Flow Control Protocol 049 Completion Timeout 050 Completer Abort 051 Unexpected Completion 052 Receiver Overflow 053 Malformed TLP 054 ECRC 055 Unsupported Request 056 ACS Violation 057 Uncorrectable Internal Error 058 MC Blocked TLP 059 AtomicOp Egress Blocked 060 TLP Prefix Blocked Error 061 TOTAL_ERR_FATAL 062 63What: /sys/bus/pci/devices/<dev>/aer_dev_nonfatal64Date: July 201865KernelVersion: 4.19.066Contact: linux-pci@vger.kernel.org, rajatja@google.com67Description: List of uncorrectable nonfatal errors seen and reported by this68 PCI device using ERR_NONFATAL. Note that since multiple errors69 may be reported using a single ERR_FATAL message, thus70 TOTAL_ERR_NONFATAL at the end of the file may not match the71 actual total of all the errors in the file. Sample output::72 73 localhost /sys/devices/pci0000:00/0000:00:1c.0 # cat aer_dev_nonfatal74 Undefined 075 Data Link Protocol 076 Surprise Down Error 077 Poisoned TLP 078 Flow Control Protocol 079 Completion Timeout 080 Completer Abort 081 Unexpected Completion 082 Receiver Overflow 083 Malformed TLP 084 ECRC 085 Unsupported Request 086 ACS Violation 087 Uncorrectable Internal Error 088 MC Blocked TLP 089 AtomicOp Egress Blocked 090 TLP Prefix Blocked Error 091 TOTAL_ERR_NONFATAL 092 93PCIe Rootport AER statistics94----------------------------95 96These attributes show up under only the rootports (or root complex event97collectors) that are AER capable. These indicate the number of error messages as98"reported to" the rootport. Please note that the rootports also transmit99(internally) the ERR_* messages for errors seen by the internal rootport PCI100device, so these counters include them and are thus cumulative of all the error101messages on the PCI hierarchy originating at that root port.102 103What: /sys/bus/pci/devices/<dev>/aer_rootport_total_err_cor104Date: July 2018105KernelVersion: 4.19.0106Contact: linux-pci@vger.kernel.org, rajatja@google.com107Description: Total number of ERR_COR messages reported to rootport.108 109What: /sys/bus/pci/devices/<dev>/aer_rootport_total_err_fatal110Date: July 2018111KernelVersion: 4.19.0112Contact: linux-pci@vger.kernel.org, rajatja@google.com113Description: Total number of ERR_FATAL messages reported to rootport.114 115What: /sys/bus/pci/devices/<dev>/aer_rootport_total_err_nonfatal116Date: July 2018117KernelVersion: 4.19.0118Contact: linux-pci@vger.kernel.org, rajatja@google.com119Description: Total number of ERR_NONFATAL messages reported to rootport.120