2019-11-25 13:23:11

by Edward Cree

Subject: Re: [PATCH v2 net-next] net: core: use listified Rx for GRO_NORMAL in napi_gro_receive()

On 25/11/2019 12:02, Alexander Lobakin wrote:
> I'm not very familiar with iwlwifi, but as a work around manual
> napi_gro_flush() you can also manually flush napi->rx_list to
> prevent packets from stalling:
>
> diff -Naur a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
> --- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c    2019-11-25 14:55:03.610355230 +0300
> +++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c    2019-11-25 14:57:29.399556868 +0300
> @@ -1526,8 +1526,16 @@
>      if (unlikely(emergency && count))
>          iwl_pcie_rxq_alloc_rbs(trans, GFP_ATOMIC, rxq);
>
> -    if (rxq->napi.poll)
> +    if (rxq->napi.poll) {
> +        if (rxq->napi.rx_count) {
> +            netif_receive_skb_list(&rxq->napi.rx_list);
> +
> +            INIT_LIST_HEAD(&rxq->napi.rx_list);
> +            rxq->napi.rx_count = 0;
> +        }
> +
>          napi_gro_flush(&rxq->napi, false);
> +    }
>
>      iwl_pcie_rxq_restock(trans, rxq);
>  }
... or we could export gro_normal_list(), instead of open-coding it
 in the driver?

-Ed

2019-11-25 13:41:01

by Alexander Lobakin

Subject: Re: [PATCH v2 net-next] net: core: use listified Rx for GRO_NORMAL in napi_gro_receive()

Edward Cree wrote 25.11.2019 16:21:
> On 25/11/2019 12:02, Alexander Lobakin wrote:
>> I'm not very familiar with iwlwifi, but as a work around manual
>> napi_gro_flush() you can also manually flush napi->rx_list to
>> prevent packets from stalling:
>>
>> diff -Naur a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
>> b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
>> --- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c    2019-11-25
>> 14:55:03.610355230 +0300
>> +++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c    2019-11-25
>> 14:57:29.399556868 +0300
>> @@ -1526,8 +1526,16 @@
>>      if (unlikely(emergency && count))
>>          iwl_pcie_rxq_alloc_rbs(trans, GFP_ATOMIC, rxq);
>>
>> -    if (rxq->napi.poll)
>> +    if (rxq->napi.poll) {
>> +        if (rxq->napi.rx_count) {
>> +            netif_receive_skb_list(&rxq->napi.rx_list);
>> +
>> +            INIT_LIST_HEAD(&rxq->napi.rx_list);
>> +            rxq->napi.rx_count = 0;
>> +        }
>> +
>>          napi_gro_flush(&rxq->napi, false);
>> +    }
>>
>>      iwl_pcie_rxq_restock(trans, rxq);
>>  }
> ... or we could export gro_normal_list(), instead of open-coding it
>  in the driver?

I thought about this too, but don't like it. This patch is proposed as
a *very* temporary solution until iwlwifi will get more straightforward
logic. I wish we could make napi_gro_flush() static in the future and
keep gro_normal_list() private to:

- prevent them from using in any new drivers;
- give more opportunity to CC to optimize the core code.

>
> -Ed

Regards,
ᚷ ᛖ ᚢ ᚦ ᚠ ᚱ