ix.4: rename and improve #1494
ix.4: rename and improve #1494
Conversation
PR: 213026 MFC after: 3 days Reported by: Sergey Akhmatov <[email protected]> Co-authored-by: Alexander Ziaee <[email protected]>
|
Instead of LOADER TUNABLES, is it appropriate to use ENVIORNMENT? We have plenty of pages with the former, but the latter is a standardized section as specified in mdoc(7). |
share/man/man4/ix.4
Outdated
| @@ -74,7 +74,8 @@ For information on enabling VLANs, see | |||
| .Sh HARDWARE | |||
| The | |||
| .Nm | |||
| driver supports the following cards: | |||
| driver supports 10Gb Ethernet PCIe adapters based on the 82598EB controller, | |||
There was a problem hiding this comment.
This change is because the entire hardware section is what appears in the hardware release notes.
There was a problem hiding this comment.
Sorry the line comment is not exact enough, the part I would remove is specifically the based on the 82598EB controller. While it is true in some absolute sense, each of these cards is marketed independently and the rest of the family is closer affined to the 82599 (aka X520) while it does not add any particular value to the documentation.
There was a problem hiding this comment.
Should we mention 82599 in DESCRIPTION? E.g. based on the Intel 82598EB and 82599 Intel(R)...
There was a problem hiding this comment.
The HARDWARE section would be more appropriate, I would not recommend any generic mention.
The HARDWARE section could be cleaned up/fleshed out a bit based on ixgbe_vendor_info_array in src/sys/dev/ixgbe/if_ix.c.. I would say something like this is accurate:
- 82598EB
- 82599 (X520) and X520 Bypass
- X540 and X540 Bypass
- X550
- X552
- X553
I will defer to someone that has a doc bit for that. |
I've been using LOADER TUNABLES or SYSCTL VARIABLRS. Environment is definitely wrong Warner |
PR: 213026 MFC after: 3 days Co-authored-by: Alexander Ziaee <[email protected]>
concussious
force-pushed
the
ix.4
branch
2 times, most recently
from
cfd9b06
to
5a7ac84
Compare
share/man/man4/ix.4
Outdated
| .It | ||
| Intel(R) Ethernet X552 (10Gb/5Gb/2.5Gb/1Gb/100Mb) | ||
| .It | ||
| Intel(R) Ethernet X553 (10Gb/5Gb/2.5Gb/1Gb/100Mb) |
There was a problem hiding this comment.
I would remove all the listed speeds because I think all of these adapters can go all the way down to 10mbit with a copper PHY and should be able to do 100mbit SERDES.
It would be helpful to document the 2.5/5Gb capable adapters including the sysctl dev.ix.0.advertise_speed to enable them in auto-negotiate but maybe in a follow up PR to move this along.
concussious
force-pushed
the
ix.4
branch
2 times, most recently
from
86783cf
to
f4f3e58
Compare
+ add controller name to title for search keywords - remove "for the freebsd operating system" from title/search keywords - remove `(R)` from the page title for consistency with other drivers - increase specificity of HARDWARE for inclusion in Release HW Notes - order HARDWARE in reverse chronological so non-eol appear first - s/PCI/PCIe/ MFC after: 3 days Reported by: kbowling (additional hardware supported)
MFC after: 3 days
|
Sorry for pushing again, there was some extra whitespace on 3 of my lines. Thanks so much, this was cool! |
+ add controller name to title for search keywords - remove "for the freebsd operating system" from title/search keywords - remove `(R)` from the page title for consistency with other drivers - increase specificity of HARDWARE for inclusion in Release HW Notes - order HARDWARE in reverse chronological so non-eol appear first - s/PCI/PCIe/ MFC after: 3 days Reported by: kbowling (additional hardware supported) Pull Request: #1494
MFC after: 3 days Pull Request: #1494
Describe better for ongoing projects to improve apropos results and the release hardware notes.
Since Felix Johnson has (seemingly?) not been seen for over 3 years, base these improvements from his (very lightly edited) proposed improvements 1:
@kev009
Footnotes
https://reviews.freebsd.org/D32531 ↩