[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <be9b4290-1927-4353-8a15-bffe84769d23@redhat.com>
Date: Tue, 5 Nov 2024 13:15:09 +0100
From: Paolo Abeni <pabeni@...hat.com>
To: chengyechun <chengyechun1@...wei.com>, netdev@...r.kernel.org,
j.vosburgh@...il.com, andy@...yhouse.net
Cc: vfalico@...il.com
Subject: Re: [PATCH] enable port after switch
On 10/31/24 03:34, chengyechun wrote:
> After switching the best aggregator,
> change the backup value of the corresponding slave node to 0
>
> Signed-off-by: chengyechun <chengyechun1@...wei.com>
>
> ---
> drivers/net/bonding/bond_3ad.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
> index b19e0e41b..b07e42950 100644
> --- a/drivers/net/bonding/bond_3ad.c
> +++ b/drivers/net/bonding/bond_3ad.c
> @@ -1830,6 +1830,8 @@ static void ad_agg_selection_logic(struct aggregator *agg,
> __disable_port(port);
> }
> }
> + port = best->lag_ports;
> + __enable_port(port);
> /* Slave array needs update. */
> *update_slave_arr = true;
> }
The above has several issues:
- does not apply
- does not include the target tree in the subj prefix ('net-next')
- the commit message does not describe why such thing should be needed
- Only the first port in the best lag group is enabled instead of all of
them, as done a few lines later - but only if there is an active partner.
Thanks,
Paolo
Powered by blists - more mailing lists