[hns3-0114]net: hns3: log and clear hardware error after reset complete

[Fred Kimmy]

[Bug Description]
When device is resetting, the CMDQ service may be stopped until
reset completed. If a new RAS error occurs at this moment, it
will no be able to clear the RAS source.

[Steps to Reproduce]
1.enject RAS & global reset at the same time

[Actual Results]
RAS error cannot be cleared

[Expected Results]
RAS error can be cleared

[Reproducibility]
Inevitably

[Additional information]
Hardware: D06
Firmware: NA
Kernel: NA

[Resolution]
This patch fixes it by clearing the RAS source after reset complete.

4fdd0bca6152 net: hns3: log and clear hardware error after reset complete

[Taihsiang Ho (tai271828)]

The upstream commit uses a function "hclge_handle_all_hns_hw_errors", and this function is not implemented in Bionic kernel source.

[Taihsiang Ho (tai271828)]

hclge_handle_all_hns_hw_errors shows up in the v5.3 upstream kernel for the first time.

4fdd0bca6152 shows up in the v5.5 upstream kernel for the first time.

However, the bionic stable channel update mainly pulls upstream patches via v4.14 and v4.19, which are the nearest upstream longterm kernel. Backporting this patch (4fdd0bca6152) will need a lot extra patches out of v4.14 and v4.19, and it will only helps logging enhancement rather than a malfunction fix. I would suggest to reduce the risk of device malfunction by "won't fix".

[Taihsiang Ho (tai271828)]

Already in bionic-hwe since Ubuntu-hwe-5.3.0-40.32_18.04.1.

[Taihsiang Ho (tai271828)]

For focal:

$ lp_master_next_tags "net: hns3: log and clear hardware error after reset complete"
33dc5e5d22dd
Ubuntu-5.4-5.4.0-10.13
Ubuntu-5.4-5.4.0-11.14
Ubuntu-5.4-5.4.0-12.15
Ubuntu-5.4-5.4.0-13.16
Ubuntu-5.4-5.4.0-14.17
Ubuntu-5.4.0-15.18
Ubuntu-5.4.0-16.19
Ubuntu-5.4.0-17.20
Ubuntu-5.4.0-17.21