On 11/13/2020 2:44 AM, Andrew Lunn wrote:
On Thu, Nov 12, 2020 at 05:54:51PM +0200, Moshe Shemesh wrote:
On 11/12/2020 3:13 PM, Andrew Lunn wrote:
On Thu, Nov 12, 2020 at 07:49:41AM +0200, Moshe Shemesh wrote:
From: Vladyslav Tarasiuk <vladysl...@nvidia.com>
DSFP is a new cable module type, which EEPROM uses memory layout
described in CMIS 4.0 document. Use corresponding standard value for
userspace ethtool to distinguish DSFP's layout from older standards.
Add DSFP module ID in accordance to SFF-8024.
DSFP module memory can be flat or paged, which is indicated by a
flat_mem bit. In first case, only page 00 is available, and in second -
multiple pages: 00h, 01h, 02h, 10h and 11h.
You are simplifying quite a bit here, listing just these pages. When i
see figure 8-1, CMIS Module Memory Map, i see many more pages, and
banks of pages.
Right, but as I understand these are the basic 5 pages which are mandatory.
Supporting more than that we will need new API.
Humm, actually, looking at the diagram again, pages 10h and 11h are
banked. Is one bamk sufficient? If so, you need to document that bank
zero is always returned, and make sure your firmware is doing that.
We also need to be clear that tunable laser information is not
available, due to this fixed layout.
Pages 10h and 11h are banked, but the mandatory data for active
transceivers is in bank 0. I will document it in the commit. There are
more banked pages, but they are all optional. Also pages 03h and 04h are
optional. I am looking to support here the mandatory part. For
supporting the optional pages we will need to change API.
Andrew
