Re: [PATCH net-next V6 2/3] net/mlx5e: Avoid copying payload to the skb's linear part

[Dragos Tatulea <dtatulea@nvidia.com>]

Hi Amery,

On 07.05.26 15:53, Amery Hung wrote:
> [...]
> Am I understanding correctly that the better performance comes with
> the assumption that the XDP does not change headers?
> 
> headlen is determined before the XDP program runs. If it push/pop
> headers, there could be headers in frags or data in the linear region
> after __pskb_pull_tail().
> 
That's right.

>>                         if (__test_and_clear_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags)) {
>>                                 struct mlx5e_frag_page *pfp;
>> @@ -2060,8 +2066,7 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w
>>                                 pagep->frags++;
>>                         while (++pagep < frag_page);
>>
>> -                       headlen = min_t(u16, MLX5E_RX_MAX_HEAD - len,
>> -                                       skb->data_len);
>> +                       headlen = min_t(u16, headlen - len, skb->data_len);
> 
> headlen - len can underflow but will be capped by skb->data_len, so
> this should be okay, right?
It is safe. But it might trigger an extra allocation in the pull when
len > headlen. We could also skip the pull in that case. Or do a
min(headlen - len, min(skb->data_len, MLX5E_RX_MAX_HEAD)). WDYT?

Thanks,
Dragos