EEH report when Network card iperf test

[lili-lilili]

When we do the netcard iperf test(on P10), the EEH error always report.

Here is the register dump from OPAL. It looks like the PCIe link get some link problems. "phbRegbErrorStatus" register records a link down error.

I also had problems with parsing the registers. I only have a pcie spec for P9, bu it has no explanation for "phbRegbErrorStatus bit31".

Any suggestions? @fbarrat

[ 1005.489389250,3] PHB#006f[6:3]: PHB Freeze/Fence detected ! [ 1005.489438608,3] PHB#006f[6:3]: PCI FIR=0000000000000000 [ 1005.489474501,3] PHB#006f[6:3]: PCI FIR WOF=0000000000000000 [ 1005.489510147,3] PHB#006f[6:3]: NEST FIR=0800000000000000 [ 1005.489542000,3] PHB#006f[6:3]: NEST FIR WOF=0800000000000000 [ 1005.489577729,3] PHB#006f[6:3]: ERR RPT0=0010000000000000 [ 1005.489607921,3] PHB#006f[6:3]: ERR RPT1=0000000000000000 [ 1005.489642726,3] PHB#006f[6:3]: AIB ERR=00cc100000000000 [ 1005.490223446,3] PHB#006f[6:3]: brdgCtl = 00000002 [ 1005.490262691,3] PHB#006f[6:3]: deviceStatus = 00000140 [ 1005.490295542,3] PHB#006f[6:3]: slotStatus = 00402000 [ 1005.490335762,3] PHB#006f[6:3]: linkStatus = c1010008 [ 1005.490366070,3] PHB#006f[6:3]: devCmdStatus = 00100107 [ 1005.490397781,3] PHB#006f[6:3]: devSecStatus = 00000000 [ 1005.490428166,3] PHB#006f[6:3]: rootErrorStatus = 00000000 [ 1005.490463922,3] PHB#006f[6:3]: corrErrorStatus = 00000000 [ 1005.490497253,3] PHB#006f[6:3]: uncorrErrorStatus = 00000000 [ 1005.490530530,3] PHB#006f[6:3]: devctl = 00000140 [ 1005.490562582,3] PHB#006f[6:3]: devStat = 00000000 [ 1005.490593271,3] PHB#006f[6:3]: tlpHdr1 = 00000000 [ 1005.490623159,3] PHB#006f[6:3]: tlpHdr2 = 00000000 [ 1005.490654178,3] PHB#006f[6:3]: tlpHdr3 = 00000000 [ 1005.490687329,3] PHB#006f[6:3]: tlpHdr4 = 00000000 [ 1005.490721063,3] PHB#006f[6:3]: sourceId = 00000000 [ 1005.490751859,3] PHB#006f[6:3]: nFir = 0800000000000000 [ 1005.490786389,3] PHB#006f[6:3]: nFirMask = 003001d000000000 [ 1005.490820066,3] PHB#006f[6:3]: nFirWOF = 0800000000000000 [ 1005.490853079,3] PHB#006f[6:3]: phbPlssr = 0000001c00000000 [ 1005.490888389,3] PHB#006f[6:3]: phbCsr = 0000001c00000000 [ 1005.490926839,3] PHB#006f[6:3]: lemFir = 0000000100000100 [ 1005.490964803,3] PHB#006f[6:3]: lemErrorMask = 0000000000000000 [ 1005.491002854,3] PHB#006f[6:3]: lemWOF = 0000000000000100 [ 1005.491036256,3] PHB#006f[6:3]: phbErrorStatus = 000004e000000000 [ 1005.491069266,3] PHB#006f[6:3]: phbFirstErrorStatus = 0000040000000000 [ 1005.491102274,3] PHB#006f[6:3]: phbErrorLog0 = 0000000000000000 [ 1005.491140034,3] PHB#006f[6:3]: phbErrorLog1 = 0000000000000000 [ 1005.491177123,3] PHB#006f[6:3]: phbTxeErrorStatus = 0000000000000000 [ 1005.491213447,3] PHB#006f[6:3]: phbTxeFirstErrorStatus = 0000000000000000 [ 1005.491248051,3] PHB#006f[6:3]: phbTxeErrorLog0 = 0000000000000000 [ 1005.491281223,3] PHB#006f[6:3]: phbTxeErrorLog1 = 0000000000000000 [ 1005.491314903,3] PHB#006f[6:3]: phbRxeArbErrorStatus = 0000000000000000 [ 1005.491349975,3] PHB#006f[6:3]: phbRxeArbFrstErrorStatus = 0000000000000000 [ 1005.491388215,3] PHB#006f[6:3]: phbRxeArbErrorLog0 = 0000000000000000 [ 1005.491425458,3] PHB#006f[6:3]: phbRxeArbErrorLog1 = 0000000000000000 [ 1005.491459174,3] PHB#006f[6:3]: phbRxeMrgErrorStatus = 0000000000000001 [ 1005.491492464,3] PHB#006f[6:3]: phbRxeMrgFrstErrorStatus = 0000000000000001 [ 1005.491524353,3] PHB#006f[6:3]: phbRxeMrgErrorLog0 = 0000000000000000 [ 1005.491557762,3] PHB#006f[6:3]: phbRxeMrgErrorLog1 = 0000000000000000 [ 1005.491594088,3] PHB#006f[6:3]: phbRxeTceErrorStatus = 0000000000000000 [ 1005.491632865,3] PHB#006f[6:3]: phbRxeTceFrstErrorStatus = 0000000000000000 [ 1005.491671010,3] PHB#006f[6:3]: phbRxeTceErrorLog0 = 0000000000000000 [ 1005.491705600,3] PHB#006f[6:3]: phbRxeTceErrorLog1 = 0000000000000000 [ 1005.491738612,3] PHB#006f[6:3]: phbPblErrorStatus = 0100000000000000 [ 1005.491772417,3] PHB#006f[6:3]: phbPblFirstErrorStatus = 0100000000000000 [ 1005.491810821,3] PHB#006f[6:3]: phbPblErrorLog0 = 0000000000000000 [ 1005.491846866,3] PHB#006f[6:3]: phbPblErrorLog1 = 0000000000000000 [ 1005.491884228,3] PHB#006f[6:3]: phbPcieDlpErrorLog1 = 0000000000000000 [ 1005.491921107,3] PHB#006f[6:3]: phbPcieDlpErrorLog2 = 0000000000000000 [ 1005.491954581,3] PHB#006f[6:3]: phbPcieDlpErrorStatus = 0000000000000000 [ 1005.491988376,3] PHB#006f[6:3]: phbRegbErrorStatus = 0090001100000000 [ 1005.492022294,3] PHB#006f[6:3]: phbRegbFirstErrorStatus = 0000001000000000 [ 1005.492058404,3] PHB#006f[6:3]: phbRegbErrorLog0 = 2480005800000000 [ 1005.492094757,3] PHB#006f[6:3]: phbRegbErrorLog1 = 0000000000000000