On Thu, 21 Jul 2022 21:22:16 +0300
Vladimir Oltean [off-list ref] wrote:

On Thu, Jul 21, 2022 at 07:21:45PM +0200, Marek Behún wrote:

quoted

Marvell documentation says that 2500base-x does not implement inband
AN.  

Does Marvell documentation actually call it 2500base-x when it says it
doesn't support in-band autoneg?

Yes, it does.

quoted

But when it was first implemented, for some reason it was thought that
2500base-x is just 1000base-x at 2.5x speed, and 1000base-x does
support inband AN. Also it worked during tests for both switches and
SOC NICs, so it was enabled.

At the time 2500base-x was not standardized. Now 2500base-x is
stanradrized, and the standard says that 2500base-x does not support
clause 37 AN. I guess this is because where it is used, it is intended
to work with clause 73 AN somehow.  

When you say 2500base-x is standardized, do you mean there is a document
somewhere which I could use to read more about this?

IEEE Std 802.3cb-2018: Amendment 1: Physical Layer Specifications and
Management Parameters for 2.5 Gb/s and 5 Gb/s Operation over Backplane.

Annex 127A (informative): Compatibility of 2.5GBASE-X PCS/PMA with
1000BASE-X PCS/PMA running 2.5 times faster

  ...
  This annex discusses the restrictions when operating 2.5GBASE-X
  PCS/PMA with a 1000BASE-X PCS/PMA link partner running 2.5 times
  faster. Compatibility of the PMD is outside the scope of this annex.
  In this annex when 1000BASE-X PCS/PMA is referred to, the 2.5 times
  speed up is implied.
  ...
  The 2.5GBASE-X PCS does not support Clause 37 Auto-Negotiation.
  Hence, the 1000BASE-X PCS is expected to have its Clause 37
  Auto-Negotiation functionality disabled so that the /C/ ordered set
  will not be transmitted. If a 2.5GBASE-X PCS receives /C/ ordered
  set, then undefined behavior may occur.
  ...

Marek