Re: [dpdk-dev] [PATCH v2 2/5] net/hns3: use unsigned types for bit operator

[oulijun <oulijun@huawei.com>]

在 2020/11/7 0:38, Ferruh Yigit 写道:
> On 11/6/2020 3:51 AM, Lijun Ou wrote:
>> From: Hongbo Zheng <zhenghongbo3@huawei.com>
>>
>> According to bit operator reliability style, variables in
>> the right expression participating int bit operation
>> cannot be of unsigned type.
> 
> Assuming this is talking about BIT() ("#define BIT(nr) (1UL << (nr))"),
> is this description says, in the "a << b", 'b' can't be unsigned?
> 
Sorry, The commit write error. according to the coverity check, it need 
to satisfies for the right expression to be unsigned.
> The code below does the opposite, "int i" -> "uint32_t i", even though 
> there is a typo in above description, why 'b' can't be signed?
> It can't be negative, but not sure if is it a problem to have it signed.
> 
> 
> Also only first change in this patch seems related to the patch title 
> and the description, rest looks related to signed / unsigned comparison 
> fixes, if so can you separate them into their patch with proper 
> description please?
> 
Yes, I put them together because they are all coveredity alarms and 
bit-operator alarms.
>>
>> Signed-off-by: Hongbo Zheng <zhenghongbo3@huawei.com>
>> Signed-off-by: Lijun Ou <oulijun@huawei.com>
>> ---
>>   drivers/net/hns3/hns3_ethdev_vf.c     |  2 +-
>>   drivers/net/hns3/hns3_rxtx_vec_neon.h | 11 +++++------
>>   2 files changed, 6 insertions(+), 7 deletions(-)
>>
>> diff --git a/drivers/net/hns3/hns3_ethdev_vf.c 
>> b/drivers/net/hns3/hns3_ethdev_vf.c
>> index 6f71cd6..2e9bfda 100644
>> --- a/drivers/net/hns3/hns3_ethdev_vf.c
>> +++ b/drivers/net/hns3/hns3_ethdev_vf.c
>> @@ -1331,7 +1331,7 @@ hns3vf_get_tc_info(struct hns3_hw *hw)
>>   {
>>       uint8_t resp_msg;
>>       int ret;
>> -    int i;
>> +    uint32_t i;
>>       ret = hns3_send_mbx_msg(hw, HNS3_MBX_GET_TCINFO, 0, NULL, 0,
>>                   true, &resp_msg, sizeof(resp_msg));
>> diff --git a/drivers/net/hns3/hns3_rxtx_vec_neon.h 
>> b/drivers/net/hns3/hns3_rxtx_vec_neon.h
>> index 8d7721b..fe525de 100644
>> --- a/drivers/net/hns3/hns3_rxtx_vec_neon.h
>> +++ b/drivers/net/hns3/hns3_rxtx_vec_neon.h
>> @@ -89,13 +89,12 @@ hns3_desc_parse_field(struct hns3_rx_queue *rxq,
>>                 struct hns3_desc *rxdp,
>>                 uint32_t   bd_vld_num)
>>   {
>> -    uint32_t l234_info, ol_info, bd_base_info;
>> +    uint32_t l234_info, ol_info, bd_base_info, cksum_err, i;
> 
> Not sure combining more variable declarations into same line is good 
> idea, why not have their own lines?
> 
Yes. I agree with you. I will fixes it in next version.
>>       struct rte_mbuf *pkt;
>>       uint32_t retcode = 0;
>> -    uint32_t cksum_err;
>> -    int ret, i;
>> +    int ret;
>> -    for (i = 0; i < (int)bd_vld_num; i++) {
>> +    for (i = 0; i < bd_vld_num; i++) {
>>           pkt = sw_ring[i].mbuf;
>>           /* init rte_mbuf.rearm_data last 64-bit */
>> @@ -129,9 +128,9 @@ hns3_recv_burst_vec(struct hns3_rx_queue 
>> *__restrict rxq,
>>       uint16_t rx_id = rxq->next_to_use;
>>       struct hns3_entry *sw_ring = &rxq->sw_ring[rx_id];
>>       struct hns3_desc *rxdp = &rxq->rx_ring[rx_id];
>> -    uint32_t bd_valid_num, parse_retcode;
>> +    uint32_t bd_valid_num, parse_retcode, pos;
>>       uint16_t nb_rx = 0;
>> -    int pos, offset;
>> +    int offset;
>>       /* mask to shuffle from desc to mbuf's rx_descriptor_fields1 */
>>       uint8x16_t shuf_desc_fields_msk = {
>>
> 
> .
>