Re: [RFC PATCH net-next 08/10] net: hibmcge: Implement workqueue and some ethtool_ops functions

From: Andrew Lunn
Date: Thu Aug 01 2024 - 16:16:25 EST

Next message: Gustavo A. R. Silva: "Re: [gustavoars:testing/wfamnae-next20240729-cbc-2 11/18] include/rdma/uverbs_ioctl.h:643:15: error: static assertion failed due to requirement '__builtin_offsetof(struct uverbs_attr_bundle, attrs) == sizeof(struct uverbs_attr_bundle_hdr)': struct member likely outside of struct_group_tagged()"
Previous message: Aruna Ramakrishna: "Re: [PATCH v7 2/5] x86/pkeys: Add helper functions to update PKRU on the sigframe"
In reply to: Jijie Shao: "Re: [RFC PATCH net-next 08/10] net: hibmcge: Implement workqueue and some ethtool_ops functions"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

> If the network port is linked, but the link fails between the SGMII and PHY,
> is there any method to find out the cause?
>
> I've had a problem with phy link but SGMII no link due to poor contact.

So a hardware design issue? Has it been solved now, or is there
shipped hardware with this problem?

I would say there is a difference between the first prototype board,
and real products. If the real product has issues, then we should try
to address it. If it is just a prototype board, i would ignore it.

One option might be to expose your PCS as a phylib PCS. Do you get
interrupts from it when it has link up? Link down? phylink will then
combine media side status with PCS status.

Andrew