Re: [PATCH net-next V2 09/11] net/mlx5e: Add support for UNREADABLE netmem page pools

[Saeed Mahameed <saeed@kernel.org>]

On 22 May 16:26, Mina Almasry wrote:
>On Thu, May 22, 2025 at 2:46 PM Tariq Toukan <tariqt@nvidia.com> wrote:
>>
>> From: Saeed Mahameed <saeedm@nvidia.com>
>>
>> On netdev_rx_queue_restart, a special type of page pool maybe expected.
>>
>> In this patch declare support for UNREADABLE netmem iov pages in the
>> pool params only when header data split shampo RQ mode is enabled, also
>> set the queue index in the page pool params struct.
>>
>> Shampo mode requirement: Without header split rx needs to peek at the data,
>> we can't do UNREADABLE_NETMEM.
>>
>> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
>> Reviewed-by: Dragos Tatulea <dtatulea@nvidia.com>
>> Signed-off-by: Cosmin Ratiu <cratiu@nvidia.com>
>> Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
>> ---
>>  drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 5 +++++
>>  1 file changed, 5 insertions(+)
>>
>> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
>> index 9e2975782a82..485b1515ace5 100644
>> --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
>> +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
>> @@ -952,6 +952,11 @@ static int mlx5e_alloc_rq(struct mlx5e_params *params,
>>                 pp_params.netdev    = rq->netdev;
>>                 pp_params.dma_dir   = rq->buff.map_dir;
>>                 pp_params.max_len   = PAGE_SIZE;
>> +               pp_params.queue_idx = rq->ix;
>> +
>> +               /* Shampo header data split allow for unreadable netmem */
>> +               if (test_bit(MLX5E_RQ_STATE_SHAMPO, &rq->state))
>> +                       pp_params.flags |= PP_FLAG_ALLOW_UNREADABLE_NETMEM;
>>
>
>This patch itself looks good to me for FWIW, but unreadable netmem
>will return netmem_address(netmem) == NULL, which from an initial look
>didn't seem like you were handling in the previous patches. Not sure
>if oversight or you are sure you're not going to have unreadable
>netmem in these code paths for some reason.

I think I explained in my other reply to the other comment, we only need to
check in one location (HW_GRO payload handling).. other paths can not
support iov netmem so we are good.