why PAC N3000 8x10G port5/port6 always Up ?

[fzhan51]

ping test all passed except port5/port6;port5/port6 shows "up" even QSFPs plugged out.

[JonWay_C_Intel]

Could you confirm the status by running below command?

cat /sys/class/fpga/intel-fpga-dev.0/intel-fpga-fme.0/spi-altera.0.auto/spi_master/spi0/spi0.0/pkvl/status

Does this recover over a power cycle?

Do you know how to replicate this issue?

[fzhan51]

cat  /sys/class/../status

0x60

[fzhan51]

rsu bmcimg 06:00.0
2020-08-16 21:46:39,075 - [[pci-address(0000:06:00.0),pci_id(0x8086,0x0b30)]] performing RSU operation
2020-08-16 21:46:39,076 - [[pci-address(0000:06:02.2),pci_id(0x8086,0x2f06)]] removing device from PCIE bus
2020-08-16 21:46:41,803 - waiting 10 seconds for boot
2020-08-16 21:46:51,813 - rescanning PCIE bus: /sys/devices/pci0000:00/pci_bus/0000:00
2020-08-16 21:46:57,814 - RSU operation complete

---"power cycle" won't work

[JonWay_C_Intel]

So, correct me if im wrong here...

rsu bmcimg --> wont recover?

power cycle --> wont recover?

Always on the same ports i.e. 0,5,6 always UP? even after rsu bmcimg or power cycle?

Can you check on another card on the same machine? How many cards are showing this problem?

[fzhan51]

yes, not recovered. yet now I got one card in my hand. it seems that c827 firmware works wrong ? any method to update it ?

[JonWay_C_Intel]

Hi, I think these are different case.

The first card, if the problem follows the card. it might be the card problem. Could you provide the card S/N?

The 2nd card, there is no upgrade to the retimer needed. What is the message that you see on the fpgainfo phy? Is the problem consistent on every power up?  

[fzhan51]

The first card:the card S/N(644c36110378)

The 2nd card:currently no more card