[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <25e5b405-09bf-40ac-8f45-5f3a0bbed8f5@gmail.com>
Date: Mon, 8 Sep 2025 15:43:25 +0300
From: Tariq Toukan <ttoukan.linux@...il.com>
To: Vitaly Kuznetsov <vkuznets@...hat.com>, Li Tian <litian@...hat.com>,
netdev@...r.kernel.org, linux-hyperv@...r.kernel.org
Cc: Saeed Mahameed <saeedm@...dia.com>, Tariq Toukan <tariqt@...dia.com>,
Mark Bloch <mbloch@...dia.com>, Leon Romanovsky <leon@...nel.org>,
Haiyang Zhang <haiyangz@...rosoft.com>,
Benjamin Poirier <bpoirier@...hat.com>, Carolina Jubran
<cjubran@...dia.com>, Shahar Shitrit <shshitrit@...dia.com>
Subject: Re: [PATCH net] net/mlx5: Not returning mlx5_link_info table when
speed is unknown
On 08/09/2025 12:58, Vitaly Kuznetsov wrote:
> Li Tian <litian@...hat.com> writes:
>
>> Because mlx5e_link_mode is sparse e.g. Azure mlx5 reports PTYS 19.
>> Do not return it when speed unless retrieved successfully.
>>
>> Fixes: 65a5d35571849 ("net/mlx5: Refactor link speed handling with mlx5_link_info struct")
++
Adding author and reviewer of offending patch.
>> Suggested-by: Vitaly Kuznetsov <vkuznets@...hat.com>
>> Signed-off-by: Li Tian <litian@...hat.com>
>> ---
>> drivers/net/ethernet/mellanox/mlx5/core/port.c | 6 +++++-
>> 1 file changed, 5 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c
>> index 2d7adf7444ba..a69c83da2542 100644
>> --- a/drivers/net/ethernet/mellanox/mlx5/core/port.c
>> +++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c
>> @@ -1170,7 +1170,11 @@ const struct mlx5_link_info *mlx5_port_ptys2info(struct mlx5_core_dev *mdev,
>> mlx5e_port_get_link_mode_info_arr(mdev, &table, &max_size,
>> force_legacy);
>> i = find_first_bit(&temp, max_size);
>> - if (i < max_size)
Keep an empty line before comment.
>> + /*
Can have the comment text starting from the first line.
>> + * mlx5e_link_mode is sparse. Check speed
>
> The array is either 'mlx5e_link_mode' or 'mlx5e_ext_link_info' but both
> have holes in them.
>
You mean 'mlx5e_link_info' and 'mlx5e_ext_link_info'.
I wouldn't say they are sparse. They have holes indeed.
>> + * is non-zero as indication of a hole.
>> + */
>> + if (i < max_size && table[i].speed)
>> return &table[i];
>>
>> return NULL;
>
Powered by blists - more mailing lists