Re: [dpdk-dev] [PATCH v2 4/5] net/hns3: check PCI config space writes

[oulijun <oulijun@huawei.com>]

在 2020/11/7 0:45, Ferruh Yigit 写道:
> On 11/6/2020 3:51 AM, Lijun Ou wrote:
>> Here adds a check for the return value when calling
>> rte_pci_write_config.
>>
>> Coverity issue: 363714
>> Fixes: cea37e513329 ("net/hns3: fix FLR reset")
>> Cc: stable@dpdk.org
>>
>> Signed-off-by: Lijun Ou <oulijun@huawei.com>
>> ---
>> V1->V2:
>> - rte_pci_wirte_config -> rte_pci_write_config
>> ---
>>   drivers/net/hns3/hns3_ethdev_vf.c | 10 +++++++---
>>   1 file changed, 7 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/net/hns3/hns3_ethdev_vf.c 
>> b/drivers/net/hns3/hns3_ethdev_vf.c
>> index 2e9bfda..2f6d91b 100644
>> --- a/drivers/net/hns3/hns3_ethdev_vf.c
>> +++ b/drivers/net/hns3/hns3_ethdev_vf.c
>> @@ -139,7 +139,7 @@ hns3vf_enable_msix(const struct rte_pci_device 
>> *device, bool op)
>>           ret = rte_pci_read_config(device, &control, sizeof(control),
>>                       (pos + PCI_MSIX_FLAGS));
>>           if (ret < 0) {
>> -            PMD_INIT_LOG(ERR, "Failed to read PCI offset 0x%x",
>> +            PMD_INIT_LOG(ERR, "failed to read PCI offset 0x%x",
> 
> Can you please drop this unrelated change from the patch?
> 
Yes. I will fix it
>>                        (pos + PCI_MSIX_FLAGS));
>>               return -ENXIO;
>>           }
>> @@ -148,8 +148,12 @@ hns3vf_enable_msix(const struct rte_pci_device 
>> *device, bool op)
>>               control |= PCI_MSIX_FLAGS_ENABLE;
>>           else
>>               control &= ~PCI_MSIX_FLAGS_ENABLE;
>> -        rte_pci_write_config(device, &control, sizeof(control),
>> -                     (pos + PCI_MSIX_FLAGS));
>> +        ret = rte_pci_write_config(device, &control, sizeof(control),
>> +                      (pos + PCI_MSIX_FLAGS));
>> +        if (ret < 0) {
>> +            PMD_INIT_LOG(ERR, "failed to write PCI offset 0x%x",
>> +                    (pos + PCI_MSIX_FLAGS));
>> +        }
>>           return 0;
>>       }
>>       return -ENXIO;
>>
> 
> .
>