netfilter-devel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Pablo Neira Ayuso <pablo@netfilter.org>
To: Arturo Borrero Gonzalez <arturo.borrero.glez@gmail.com>
Cc: netfilter-devel@vger.kernel.org
Subject: Re: [iptables PATCH] nft-arp: fix inversion of matches
Date: Mon, 10 Nov 2014 18:30:29 +0100	[thread overview]
Message-ID: <20141110173029.GA29762@salvia> (raw)
In-Reply-To: <20141108213434.27991.30805.stgit@nfdev.cica.es>

On Sat, Nov 08, 2014 at 10:35:49PM +0100, Arturo Borrero Gonzalez wrote:
> Inversion of matches is failing because NFT_CMP_EQ is used unconditionally.
> 
> The family agnostic functions don't need this fix, because arp inv flags
> are translated to ipt inv flags, and these flags are well handled there.
> 
> Signed-off-by: Arturo Borrero Gonzalez <arturo.borrero.glez@gmail.com>
> ---
> NOTES: This patch is for the master branch of iptables tree.
>        Compile-tested only. Please comment.

Please, no compile-tested patches, submit RFC or tested stuff.

>  iptables/nft-arp.c |   41 ++++++++++++++++++++++++++++++++++++-----
>  1 file changed, 36 insertions(+), 5 deletions(-)
> 
> diff --git a/iptables/nft-arp.c b/iptables/nft-arp.c
> index f45ad0f..cb3623d 100644
> --- a/iptables/nft-arp.c
> +++ b/iptables/nft-arp.c
> @@ -164,6 +164,7 @@ static int nft_arp_add(struct nft_rule *r, void *data)
>  	struct arptables_command_state *cs = data;
>  	struct arpt_entry *fw = &cs->fw;
>  	uint8_t flags = arpt_to_ipt_flags(fw->arp.invflags);

I prefer if you kill arpt_to_ipt_flags(), so add_iniface() uses
native NFT_CMP_{EQ,NEQ} as third parameter. Thus, add_iniface() will
be consistent with other existing add_cmp_*() functions.

This will require some extra code in nft-ipv4 and nft-ipv6 though, but
I'd like to avoid this arpt to ipt flags conversion.

> +	uint32_t op = NFT_CMP_EQ;
>  	int ret = 0;
>  
>  	if (fw->arp.iniface[0] != '\0')
> @@ -174,12 +175,24 @@ static int nft_arp_add(struct nft_rule *r, void *data)
>  
>  	if (fw->arp.arhrd != 0) {
>  		add_payload(r, offsetof(struct arphdr, ar_hrd), 2);
> -		add_cmp_u16(r, fw->arp.arhrd, NFT_CMP_EQ);
> +
> +		if (fw->arp.invflags & ARPT_INV_ARPHRD)
> +			op = NFT_CMP_NEQ;
> +		else
> +			op = NFT_CMP_EQ;
> +
> +		add_cmp_u16(r, fw->arp.arhrd, op);

This is a good example of what I would like to see in the family
specific nft code, use this pattern consistently in this code. Thanks.

      reply	other threads:[~2014-11-10 17:28 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-11-08 21:35 [iptables PATCH] nft-arp: fix inversion of matches Arturo Borrero Gonzalez
2014-11-10 17:30 ` Pablo Neira Ayuso [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20141110173029.GA29762@salvia \
    --to=pablo@netfilter.org \
    --cc=arturo.borrero.glez@gmail.com \
    --cc=netfilter-devel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).