Re: [PATCH 1/3] libnl: add netfilter support

[Patrick McHardy <kaber@trash.net>]

Thomas Graf wrote:
> * Patrick McHardy <kaber@trash.net> 2007-09-03 12:06
> 
>>Thomas Graf wrote:
>>
>>>I think I can live with checking for the bit on the receiving side,
>>>the only problem it could cause is when the attribute type is used
>>>to encode information such as a port number. So if we are willing
>>>to take that risk the receiving path could be merged as well.
>>
>>
>>That kind of information should be stored as attribute value, no?
> 
> 
> Yes it should but we've already got code like this:
> for (i = 0; i < cnt; i++)
>         nla_put(skb, i, ....);


That looks pretty much like a list of attributes, but
using different attribute types. Just out of interest:
do you have a pointer to code doing this?


> It's only a matter of time until someone comes up with
> something like this:
> nla_put(skb, dstport, sizeof(port_info), &port_info);
> 
> Maybe we could put a WARN_ON(bit_set) in nla_put() on kernel side
> to make sure such behaviour is not being introduced by accident.


That sounds like a good idea. Anyone doing this is looking
for trouble since array-based attribute parsing would require
at least 2^31 elements :)