[illumos-Developer] [PATCH]: correct inet_pton behavior

[Garrett D'Amore]

In agree wholeheartedly.  Also, this is not a documented kernel api, and it really should only be used for logging.  I think fixing it now is the right thing to do.

  - Garrett

Vitaliy Gusev <gusev.vitaliy at nexenta.com> wrote:

>On 07/16/2011 01:09 AM, Richard Lowe wrote:
>> I'm surprised that the consumers who worked around this didn't fix it,
>> but that also makes me somewhat suspicious.
>>
>> What are the chances that 3rd party modules also work around this, and
>> would break if we fixed it?
>
>It can be a problem. But:
>
>    1. This is a bug. At least function with that name *SHOULD* return 
>result in network byte order.
>
>    2. This function is not aften used. In the kernel we see only 3 places.
>
>    3. Any caller needs to do additional steps to place result in 
>network order and those steps are different for AF_INET and AF_INET6.
>
>
>I'm thinking specifically of things like
>> AFS here.
>>
>> -- Rich
>
>
>_______________________________________________
>Developer mailing list
>Developer at lists.illumos.org
>http://lists.illumos.org/m/listinfo/developer