Hi Russell,

On 11/18/21 8:31 AM, Russell King (Oracle) wrote:

On Thu, Nov 18, 2021 at 01:59:28PM +0100, Horatiu Vultur wrote:

quoted

The 11/18/2021 09:59, Russell King (Oracle) wrote:

quoted

Another approach would be to split phylink_mii_c22_pcs_decode_state()
so that the appropriate decode function is selected depending on the
interface state, which may be a better idea.

I have tried to look for phylink_mii_c22_pcs_decode_state() and I
have found it only here [1], and seems that it depends on [2]. But not
much activity happened to these series since October.
Do you think they will still get in?

I don't see any reason the first two patches should not be sent. I'm
carrying the second one locally because I use it in some changes I've
made to the mv88e6xxx code - as I mentioned in the patchwork entry you
linked to. See:

  http://git.armlinux.org.uk/cgit/linux-arm.git/log/?h=net-queue

  "net: phylink: Add helpers for c22 registers without MDIO"

Although I notice I committed it to my tree with the wrong author. :(

Sean, please can you submit the mdiodev patch and this patch for
net-next as they have general utility? Thanks.

The mdiodev patch is already in the tree as 0ebecb2644c8 ("net: mdio:
Add helper functions for accessing MDIO devices"). The c22 patch is
submitted as [1].

--Sean

[1] https://lore.kernel.org/netdev/20211022160959.3350916-1-sean.anderson@seco.com/ (local)