On 2024/10/19 0:41, Stephen Hemminger wrote:
> On Fri, 18 Oct 2024 14:19:41 +0800
> Jie Hai <haij...@huawei.com> wrote:
>
>> @@ -306,11 +324,16 @@ hns3_parse_devargs(struct rte_eth_dev *dev)
>> &hns3_parse_dev_caps_mask, &dev_caps_mask);
>> (void)rte_kvargs_process(kvlist, HNS3_DEVARG_MBX_TIME_LIMIT_MS,
>> &hns3_parse_mbx_time_limit, &mbx_time_limit_ms);
>> - if (!hns->is_vf)
>> + if (!hns->is_vf) {
>> (void)rte_kvargs_process(kvlist,
>> HNS3_DEVARG_FDIR_VALN_MATCH_MODE,
>> &hns3_parse_vlan_match_mode,
>> &hns->pf.fdir.vlan_match_mode);
>> + (void)rte_kvargs_process(kvlist,
>> + HNS3_DEVARG_FDIR_TUPLE_CONFIG,
>> + &hns3_parse_fdir_tuple_config,
>> + &hns->pf.fdir.tuple_cfg);
>> + }
>>
>
> Side question, why the casts to void here?
> Is there some tool that is complaining (like old lint), or is there
> some error handling missing.
Yes, it will output a code check warning if don't add (void).
In our design, there will be a warning trace if user input an invalid option
(and nothing changed),
so it is OK for don't process the function's return value.
