The Board Management software support board diagnostics. Please issue dis_diag to review the state of the card(s). If you have more than one card (host or target), please append '-a X' to specify which card you want to look at, where X is the adapter number - 0, 1, 2 or 3. These are assigned in sequence when the drivers are configured. Up to 4 cards is currently supported.
# dis_diag -a 0 (for the first card in a multi-card system) -----> Dis_diag will provide some information on the card; Adapter 0 > Type : MXH932 NodeId : 4 Serial number : MXH932-CC-000015 MXH chip family : Microchip - PFX GEN4 MXH chip vendorId : 0x11f8 MXH chip device : 0x4036 MXH chip revision : 0x0 (ZA) EEPROM version : 08 EEPROM vendor info : 0x0000 Firmware version : 1.4 Card revision : CC ...the slot-interface; Clock mode slot : Global Clock mode link : Global PCIe slot state : x16, Gen3 (8 GT/s) PCIe slot capabilities : x16, Gen3 (8 GT/s) ...and the status of the link from the card; Link 0 uptime : 84734 seconds Link 0 state : ENABLED Link 0 state : x16, Gen3 (8 GT/s) Link 0 required : x16, Gen3 (8 GT/s) Link 0 capabilities : x16, Gen3 (8 GT/s) ...If this host-card is connected to a supported target-card, dis_diag (-a 1) for that card would then show the target card; Adapter 1 > Type : MXH932 Mode : TARGET ...and show the cable-link as Link 0, and then the state of its slot as the slot; PCIe slot state : x8, Gen4 (16 GT/s) PCIe slot capabilities : x16, Gen4 (16 GT/s) (...) Link 0 state : x16, Gen4 (16 GT/s) Link 0 required : x16, Gen4 (16 GT/s) Link 0 capabilities : x16, Gen4 (16 GT/s) ...as above.
The slot and link-states can be checked to verify that the transparent connection is operating as expected. Dis_diag can also extract PCIe event-counters from the supported devices;
# dis_diag -a 0 -V 9 No of uncorrectable slot error events ......... : 0 No of correctable slot error events ........... : 0 (...) No of uncorrectable link error events ......... : 0 No of correctable link error events .......... : 0 ...and details on these; Uncorrectable error cnt - PCIe Cable Link 0 ... : 0 dlperr cnt ................................ : 0 sdoenerr cnt .............................. : 0 poisoned cnt .............................. : 0 fcperr cnt ................................ : 0 (etc)
To reset the PCIe error counters, please run dis_diag with the option -clear.
# dis_diag -clear ================================================================================ Dolphin diagnostic tool -- dis_diag version 5.14.0 (Fri Aug 9 13:17:10 CEST 2019) ================================================================================ dis_diag compiled in 64 bit mode Driver : Dolphin IRM (GX) 5.14.0-d Jul 28th 2019 (rev f8ba7b5) Date : on. 14. aug. 14:54:07 +0200 2019 System : Linux Xrox 3.10.0-957.12.2.el7.x86_64 #1 SMP Tue May 14 21:24:32 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux Number of configured local adapters found: 1 Interrupt statistics is cleared for local adapter 0 AER Error statistics is cleared for local adapter 0 Reset reason statistics is cleared for local adapter 0 Reset history statistics is cleared for local adapter 0 No Diagnostic performed using this command line option #