Bug 2137643 - [RFE] Add support for ingress policing above 34 Gbps
Summary: [RFE] Add support for ingress policing above 34 Gbps
Keywords:
Status: POST
Alias: None
Product: Red Hat Enterprise Linux Fast Datapath
Classification: Red Hat
Component: openvswitch3.0
Version: FDP 22.L
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
: ---
Assignee: Adrián Moreno
QA Contact: liting
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2022-10-25 17:46 UTC by Ilya Maximets
Modified: 2023-07-31 16:48 UTC (History)
4 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed:
Target Upstream Version:
Embargoed:

Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Issue Tracker FD-2404 0 None None None 2022-10-25 17:55:07 UTC

Description Ilya Maximets 2022-10-25 17:46:15 UTC 
OVS ingress policing is implemented via act_police tc action.
Similarly to HTB in BZ 2137619, act_police is getting the main
rate/ceil configuration via 'struct tc_ratespec'.
This structure holds a 32bit rate field that reflects bytes per
second.  So, the maximum value is ~34 G bits per second.

Again, similarly to HTB, act_police has dedicated netlink
attributes TCA_POLICE_RATE64 and TCA_POLICE_PEAKRATE64 that
can be used to break the 32bit barrier.

It makes sense to add support for these attributes to be able
to configure higher speeds since modern NICs can be much faster
than the current limit.
Note You need to log in before you can comment on or make changes to this bug.