netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Hangbin Liu <liuhangbin@gmail.com>
To: Jay Vosburgh <jv@jvosburgh.net>
Cc: Andrew Lunn <andrew+netdev@lunn.ch>,
	"David S. Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
	Jiri Pirko <jiri@resnulli.us>,
	netdev@vger.kernel.org
Subject: Re: [Bonding Draft Proposal] Add lacp_prio Support for ad_select?
Date: Wed, 2 Jul 2025 07:44:30 +0000	[thread overview]
Message-ID: <aGTjXpYYXIMfl9N6@fedora> (raw)
In-Reply-To: <2946319.1751389736@famine>

Hi Jay,
On Tue, Jul 01, 2025 at 10:08:56AM -0700, Jay Vosburgh wrote:
> 	It looks like lacp_find_new_agg_lead() runs though all of the
> ports in all of the aggregators and chooses the aggregator with the
> "best" port of all.

Yes, based on the ad_select policy.

> 
> 	One downside if we were to adapt this logic or something similar
> to bonding is that there's currently no way to set the Port Priority of
> interfaces in the bond.  There is a "prio" that can be set via ip set
> ... type bond_slave prio X, which is IFLA_BOND_SLAVE_PRIO, but that's a
> failover priority, not the LACP Port Priority.

How about adding a similar parameter, e.g., ad_actor_port_prio?
Currently, the actor port priority is initialized directly as 0xFF.
We could introduce a per-port ad_actor_port_prio to be used in the
ad_select policy.

I understand that, according to the IEEE standard, port priority is used to
select the best port among multiple ports within a single aggregator.
However, since the IEEE standard doesn't define how to select between two
aggregators, we could repurpose this value similarly to how the bandwidth
and count options work in the current ad_select policy.

> 
> 	So right now, if the above logic were put into bonding, the
> local selection criteria would end up based entirely on the port number,
> which isn't configurable, and so doesn't seem especially better than
> what we have now.

[...]
> 
> 	From the above, I suspect we'll have to add some additional
> configuration parameters somewhere.  It would be nice if the System
> Priority were configurable on a per-aggregator basis, but that seems
> complicated from a UI perspective (other than something like a mapping
> of agg ID to system prio).

Thanks
Hangbin

  reply	other threads:[~2025-07-02  7:44 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-06-24  6:53 [Bonding Draft Proposal] Add lacp_prio Support for ad_select? Hangbin Liu
2025-06-26 23:28 ` Jay Vosburgh
2025-06-27  4:33   ` Hangbin Liu
2025-07-01 17:08     ` Jay Vosburgh
2025-07-02  7:44       ` Hangbin Liu [this message]
2025-07-03 19:42         ` Jay Vosburgh

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=aGTjXpYYXIMfl9N6@fedora \
    --to=liuhangbin@gmail.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=jiri@resnulli.us \
    --cc=jv@jvosburgh.net \
    --cc=kuba@kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).