{{ message }}
Nvidia/Mellanox expose ROCE ECN information on sysfs on the path #695
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
/sys/class/net/<interface>/ecn/<protocol>/ There are 2 protocols Reaction Point (rp) and Notification point (np) For each of the protocols they have a list of attributes: /sys/class/net/<interface>/ecn/<protocol>/params/<requested attribute> Each protocol will also if ECN is enabled per priority (where X is the priority): /sys/class/net/<interface>/ecn/<protocol>/enable/X This is documented here https://docs.nvidia.com/networking/display/mlnxofedv571020/explicit+congestion+notification+(ecn) The attributes are documented here: https://enterprise-support.nvidia.com/s/article/dcqcn-parameters Signed-off-by: Diego Asturias <dasturias@arista.com>
d3d9716 to
cda72fe
Compare
|
LGTM in general but it makes me wonder where to draw the line what vendor specific stuff to include here and what not.. I feel this is probably relevant enough to be included but not sure. @SuperQ wdyt? |
|
Just to be clear this is just nvidia specific because other vendors aren't exposing these values through Sysfs. these are pretty generic values for ROCEv2. For example https://docs.broadcom.com/doc/NCC-WP1XX has information on broadcom's congestion control, which also uses ECN. Ideally other vendors would also expose these values in a sysfs path instead of relying on propriety utilities. I'm not aware if other vendors that implement rocev2 plan to in the future integrate with sysfs (if they already have), but this seems fairly generic in terms of ROCEv2 |
|
@dasturiasArista Thanks for clarifying. I think this is a good argument for including it here. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Nvidia/Mellanox expose ROCE ECN information on sysfs on the path
/sys/class/net/<interface>/ecn/<protocol>/
There are 2 protocols Reaction Point (rp) and Notification point (np)
For each of the protocols they have a list of attributes:
/sys/class/net/<interface>/ecn/<protocol>/params/<requested attribute>
Each protocol will also if ECN is enabled per priority (where X is the
priority):
/sys/class/net/<interface>/ecn/<protocol>/enable/X
This is documented here
https://docs.nvidia.com/networking/display/mlnxofedv571020/explicit+congestion+notification+(ecn)
The attributes are documented here:
https://enterprise-support.nvidia.com/s/article/dcqcn-parameters