Re: [PATCH net-next] net/mlx5e: Report link down on administrative close

[manjunath.b.patil@oracle.com]

On 8/19/26 10:35 AM, manjunath.b.patil@oracle.com wrote:
> 
> On 6/11/26 1:13 PM, manjunath.b.patil@oracle.com wrote:
>> On 6/10/26 5:06 AM, Tariq Toukan wrote:
>>>
>>>
>>> On 09/06/2026 2:42, Manjunath Patil wrote:
>>>> mlx5e_update_carrier() reports both link-up and link-down carrier
>>>> changes, but an administrative down does not reach it in practice. The
>>>> close path first changes the port admin state and then clears
>>>> MLX5E_STATE_OPENED and drops carrier silently in mlx5e_close_locked().
>>>> Any queued carrier worker will skip update_carrier() once the device is
>>>> no longer opened.
>>>>
>>>> This leaves "ip link set dev <dev> down" without a matching netdev
>>>> "Link down" message, while reopening the device still reports "Link 
>>>> up".
>>>>
>>>> Report the link-down transition in mlx5e_close() before the common 
>>>> close
>>>> helper clears the opened state and drops carrier. Guard the message 
>>>> with
>>>> the current opened and carrier state to avoid duplicates when the 
>>>> netdev
>>>> is already closed or carrier is already down.
>>>>
>>>> Assisted-by: Codex:gpt-5
>>>> Signed-off-by: Manjunath Patil <manjunath.b.patil@oracle.com>
>>>> ---
>>>> Validation:
>>>> - Built an OL8 mainline test kernel from this change.
>>>> - Booted 7.1.0-rc6.bug123456.el8.v1.x86_64 on an mlx5-backed VM.
>>>> - Confirmed `ip link set dev re0 down/up` and `re1 down/up` now emit
>>>>    netdev `Link down` and `Link up` messages, alongside the existing 
>>>> RDMA
>>>>    port state notifications.
>>>>
>>>>   drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 3 +++
>>>>   1 file changed, 3 insertions(+)
>>>>
>>>> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/ 
>>>> drivers/net/ethernet/mellanox/mlx5/core/en_main.c
>>>> index 8f2b3abe0092..a04a89f0eddf 100644
>>>> --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
>>>> +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
>>>> @@ -3628,6 +3628,9 @@ int mlx5e_close(struct net_device *netdev)
>>>>       mutex_lock(&priv->state_lock);
>>>>       mlx5e_modify_admin_state(priv->mdev, MLX5_PORT_DOWN);
>>>> +    if (test_bit(MLX5E_STATE_OPENED, &priv->state) &&
>>>> +        netif_carrier_ok(netdev))
>>>> +        netdev_info(netdev, "Link down\n");
>>>>       err = mlx5e_close_locked(netdev);
>>>>       mutex_unlock(&priv->state_lock);
>>>>
>>>> base-commit: e43ffb69e0438cddd72aaa30898b4dc446f664f8
>>>
>>> Thanks for your patch.
>>>
>>> I wouldn't print "Link down" as part of the "close" callback. I'd 
>>> rather get it printed in the event handler. Currently there is a 
>>> print there, but the callback is masked by the OPENED bit.
>>>
>>> This made me revisit this area, and it surely needs some more 
>>> interesting enhancements.
>>> I'll probably come up with some changes soon.
>>
>> thank for you taking a look at this patch.
>> I will hold on and wait for your changes.
> 
> Any update on this?
> 
> I checked a v7.2-rc6 based kernel. That still showed the old behavior - 
> 'fup re0' reports 're0: Link up', while a successful 'ifdown re0' 
> remains silent.
> 
> -Manjunath

Hi Tariq,

Any update on this? Should we put in the proposed change until you come 
up better solution?

-Thanks,
Manjunath
> 
>>
>> -Manjunath
>