Re: [PATCH net] net/mlx5: Attach late uplink netdev to loaded representors

[Jianbo Liu]

On 10/13/2026 7:23 PM, Paolo Abeni wrote:
> On 10/6/26 13:02, Tariq Toukan wrote:
> > From: Jianbo Liu <jianbol@xxxxxxxxxx>
> >
> > mlx5e_vport_uplink_rep_load() returns success when the uplink netdev is
> > missing, so the ETH rep is marked loaded with a NULL netdev, the IB rep
> > copies that NULL association, and nothing repairs it later. Unbind
> > mlx5_core.eth.<N> in legacy mode, enter switchdev, bind it again, and
> > rdma link show still reports no netdev on the uplink port. The netdev
> > also keeps the NIC profile instead of the uplink rep profile.
> >
> > Add an attach_uplink_netdev rep op, run from the existing reload-reps
> > work once mlx5_core_uplink_netdev_set() reports a netdev. ETH re-runs
> > the uplink load, switching profile and setting rpriv->netdev; IB redoes
> > ib_device_set_netdev(). Deferring to the work keeps the devlink lock
> > out of probe context.
> >
> > Fixes: 6b4be64fd9fe ("net/mlx5e: Harden uplink netdev access against 
> > device unbind")
> > Signed-off-by: Jianbo Liu <jianbol@xxxxxxxxxx>
> > Reviewed-by: Shay Drori <shayd@xxxxxxxxxx>
> > Signed-off-by: Tariq Toukan <tariqt@xxxxxxxxxx>
> The AI feedback looks relevant, also this looks quite too big/dangerous
> for net at this point of the cycle.
Agreed. The missing uplink netdev is a corner case rather than a recent
regression, so I will retarget v2 to net-next. It also relies on
mlx5_esw_add_work(), added this cycle in 2a110ee54e89 ("net/mlx5:
E-Switch, introduce generic work queue dispatch helper"), which is newer
than the Fixes commit, so it would not backport to stable as is.

I will address the Sashiko bot comments in v2.

Thanks,
Jianbo