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: [nf_tables PATCH] netfilter: nf_tables: invert chain deletion abort path
Date: Thu, 14 Apr 2016 12:24:34 +0200	[thread overview]
Message-ID: <20160414102434.GA2905@salvia> (raw)
In-Reply-To: <146011296985.3580.3314850969369156279.stgit@nfdev2.cica.es>

On Fri, Apr 08, 2016 at 12:56:10PM +0200, Arturo Borrero Gonzalez wrote:
> Before this patch, chain deletetion abort path re-add chains in reverse
> order of what was originally in the ruleset.
> Invert the order, so the ruleset is exactly the same after abort.
> 
> Example, using 2 config files:
> 
> ruleset_good.nft:
> ==== 8< ====
> flush ruleset
> table ip t {
> 	chain c1 {
> 	}
> 	chain c2 {
> 	}
> 	chain c3 {
> 	}
> }
> ==== 8< ====
> 
> ruleset_bad.nft:
> ==== 8< ====
> flush ruleset
> table ip t {
> 	chain c1 {
> 	}
> 	chain c2 {
> 		jump c6
> 	}
> 	chain c3 {
> 	}
> }
> ==== 8< ====
> 
> 
> before this patch:
> 
> % nft -f ruleset_good.nft
> % nft -f ruleset_bad.nft
> % nft list ruleset
> table ip t {
> 	chain c3 {
> 	}
> 
> 	chain c2 {
> 	}
> 
> 	chain c1 {
> 	}
> }
> 
> [ note, inverse order of chain listing ]
> 
> after this patch:
> 
> % nft -f ruleset_good.nft
> % nft -f ruleset_bad.nft
> % nft list ruleset
> table ip t {
> 	chain c1 {
> 	}
> 
> 	chain c2 {
> 	}
> 
> 	chain c3 {
> 	}
> }
> 
> [ note, same order of chain listing ]
> 
> Signed-off-by: Arturo Borrero Gonzalez <arturo.borrero.glez@gmail.com>
> ---
>  net/netfilter/nf_tables_api.c |    4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
> index 2011977..8578cc6 100644
> --- a/net/netfilter/nf_tables_api.c
> +++ b/net/netfilter/nf_tables_api.c
> @@ -4060,8 +4060,8 @@ static int nf_tables_abort(struct net *net, struct sk_buff *skb)
>  			break;
>  		case NFT_MSG_DELCHAIN:
>  			trans->ctx.table->use++;
> -			list_add_tail_rcu(&trans->ctx.chain->list,
> -					  &trans->ctx.table->chains);
> +			list_add_rcu(&trans->ctx.chain->list,
> +				     &trans->ctx.table->chains);

Thanks for coming up with this Arturo.

I have a better way to fix this by not adding/removing the objects to
the lists.

Ping me back if I don't come up with the fix anytime soon.

  reply	other threads:[~2016-04-14 10:24 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-04-08 10:56 [nf_tables PATCH] netfilter: nf_tables: invert chain deletion abort path Arturo Borrero Gonzalez
2016-04-14 10:24 ` Pablo Neira Ayuso [this message]
2016-04-28  8:55   ` Arturo Borrero Gonzalez
2016-06-22 12:45   ` Pablo Neira Ayuso

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=20160414102434.GA2905@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).