On 22/03/2022 15:36, David Marchand wrote:
On Fri, Mar 11, 2022 at 6:06 PM Kevin Traynor <ktray...@redhat.com> wrote:
There are currently some checks for cross-numa polling cases to
ensure that they won't effect the accuracy of the PMD ALB.
If an rxq is pinned to a pmd core by the user it will not be
I prefer "PMD thread" for consistency (both in commitlog and code comments).
I changed to "PMD thread core" across commits. I still want to reference
core as in this case the comments are for code that has core_id and
numa_id of that core. Also for the commitlog, I think people may
resonate more with the concept of pinning an rxq to a core.
reassigned by OVS, so even if it is non-local numa polled it
will not impact PMD ALB accuracy.
To establish this, a check was made on whether the pmd core was
isolated or not. However, since other_config:pmd-rxq-isolate was
introduced, rxqs may be pinned but the pmd core not isolated.
It means that by setting pmd-rxq-isolate=false and doing non-local
numa pinning, PMD ALB may not run where it should.
If the core is isolated we can skip individual rxq checks but if not,
we should check the individual rxqs for pinning before we disallow
PMD ALB.
Also, update function comments to make it's operation clearer.
Fixes: 6193e03267c1 ("dpif-netdev: Allow pin rxq and non-isolate PMD.")
But, in any case, good catch...
Signed-off-by: Kevin Traynor <ktray...@redhat.com>
Acked-by: David Marchand <david.march...@redhat.com>
_______________________________________________
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
