Re: [net-next PATCH v4 1/3] sysctl: refactor integer handling proc code

[Cong Wang <amwang@redhat.com>]

Octavian Purdila wrote:
> On Tuesday 16 February 2010 15:08:23 you wrote:
>> Octavian Purdila wrote:
>>> On Tuesday 16 February 2010 10:41:07 you wrote:
>>>>> +
>>>>> +     if (!write && !first && left && !err)
>>>>> +             err = proc_put_newline(&buffer, &left);
>>>>> +     if (write && !err)
>>>>> +             err = proc_skip_wspace(&buffer, &left);
>>>>> +     if (err == -EFAULT /* do we really need to check for -EFAULT? */
>>>>> || +         (write && first))
>>>>> +             return err ? : -EINVAL;
>>>> The logic here seems messy, adding one or two goto's may help?
>>> OK, I'll give it a try.
>>>
>>> What about the EFAULT check, is that really required?
>> I think so, it means to keep the errno to user-space when it is EFAULT,
>> right? This seems reasonable.
>>
> 
> The problem I see is that this way we don't actually acknowledge some of the 
> set values, e.g. say that we have buffer="1 2 3" and length = 100. Although we 
> do accept values 1, 2 and 3 we don't acknowledge that to the user (as we would 
> do for, say "1 2 3 4a"), but return -EFAULT.
> 
> I think it would be better to skip this check. That means that the user will 
> get the ack for the 1, 2 and 3 values and next time it continues the write it 
> will get -EFAULT.
> 
> This will of course change the userspace ABI, albeit in a minor way, and it is 
> not clear to me if doing this is allowed (even if this new approach would be 
> the correct one).
> 

I think the right behavior is accept "1 2 3" and return the number of
bytes that we accept.