Re: [PATCH nf-next 4/4] netfilter: nf_tables: include table and chain name when dumping hooks

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Sun, May 23, 2021 at 11:26:46PM +0200, Florian Westphal wrote:
> Pablo Neira Ayuso <pablo@xxxxxxxxxxxxx> wrote:
> > On Sun, May 23, 2021 at 08:54:31PM +0200, Florian Westphal wrote:
> > > Pablo Neira Ayuso <pablo@xxxxxxxxxxxxx> wrote:
> > > > On Fri, May 21, 2021 at 01:39:22PM +0200, Florian Westphal wrote:
> > > > >   * @NFTA_HOOK_FUNCTION_NAME: hook function name (NLA_STRING)
> > > > >   * @NFTA_HOOK_MODULE_NAME: kernel module that registered this hook (NLA_STRING)
> > > > > + * @NFTA_HOOK_NFT_CHAIN_INFO: nft chain and table name (NLA_NESTED)
> > > > 
> > > > Probably NFTA_HOOK_CHAIN_INFO ?
> > > 
> > > I added _NFT_ to avoid ambiguity in case this would be extended
> > > to add xt-legacy chain info.  I can drop the _NFT_, let me know.
> > 
> > It's a NLA_NESTED, you might add a _TYPE field inside the nest to
> > describe what type of chain info is stored there, maybe?
> 
> It uses enum nft_chain_attributes, it somehow feels wrong to add a
> 'type' field for that.

Agreed. Probably another nest level.

NFTA_HOOK_CHAIN_INFO
    CHAIN_INFO_DESC
         nft_chain_attributes
    CHAIN_INFO_TYPE

> I could add a new enum if you prefer.
> 
> At this point I don't think adding xt specific info is useful
> because the chain function name already tell if its mangle, raw etc.

I'd prefer to not expose internal kernel functions names, but I
understand this approach is simple.



[Index of Archives]     [Netfitler Users]     [Berkeley Packet Filter]     [LARTC]     [Bugtraq]     [Yosemite Forum]

  Powered by Linux