From: Florian Westphal <fw@strlen.de>
To: Pablo Neira Ayuso <pablo@netfilter.org>
Cc: Florian Westphal <fw@strlen.de>,
netfilter-devel <netfilter-devel@vger.kernel.org>
Subject: Re: [PATCH 0/3] netfilter: add connlabel conntrack extension
Date: Fri, 16 Nov 2012 12:31:09 +0100 [thread overview]
Message-ID: <20121116113109.GK20678@breakpoint.cc> (raw)
In-Reply-To: <20121116100256.GA25964@1984>
Pablo Neira Ayuso <pablo@netfilter.org> wrote:
> On Thu, Nov 15, 2012 at 04:55:12PM +0100, Florian Westphal wrote:
> > Changes since RFCv2:
> > - make it a variable-size extension and remove dynamic
> > reallocation of the label array
> > - add ctnetlink support for receiving/setting labels
> > - limit to 128 instead of 1k labels due to limited extension
> > space (128 is more than enough for now, so this is no problem).
>
> At a quick glance I like this round, you got this simplified :-).
>
> My only concern is that (if I'm not missing anything) it
> inconditionally add the extension even if we don't need it.
Its conditional:
words = ACCESS_ONCE(net->ct.label_words);
if (words == 0 || WARN_ON_ONCE(words > 8))
return NULL;
cl_ext = nf_ct_ext_add_length(ct, NF_CT_EXT_LABELS, words * sizeof(long),..
xt_connlabel increases "label_words" to a non-zero
value based on the highest bit it saw so far. So e.g. when bit 32 is
requested, a 32-bit machine kernel will set words to 2, and
a 64 bit kernel to 1, thus turning the extension on for new connections.
The only drawback is that when you add a connlabel rule for bits 127,
then for bit 1, then delete the first rule again words will be larger
than one.
But i think this ok (I don't want to add per-bit refcountr :-) )
next prev parent reply other threads:[~2012-11-16 11:31 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-11-15 15:55 [PATCH 0/3] netfilter: add connlabel conntrack extension Florian Westphal
2012-11-15 15:55 ` [PATCH 1/3] " Florian Westphal
2012-11-15 15:55 ` [PATCH 2/3] netfilter: ctnetlink: deliver labels to userspace via CTA_LABELS attribute Florian Westphal
2012-11-15 15:55 ` [PATCH 3/3] netfilter: ctnetlink: allow userspace to set labels Florian Westphal
2012-11-27 11:18 ` Pablo Neira Ayuso
2012-11-27 11:50 ` Florian Westphal
2012-11-27 12:31 ` Pablo Neira Ayuso
2012-11-27 13:09 ` Florian Westphal
2012-11-27 14:13 ` Pablo Neira Ayuso
2012-11-27 14:24 ` Florian Westphal
2012-11-30 13:58 ` Pablo Neira Ayuso
2012-11-30 14:02 ` Florian Westphal
2012-11-30 18:34 ` Pablo Neira Ayuso
2012-11-30 21:36 ` Florian Westphal
2012-12-03 11:04 ` Pablo Neira Ayuso
2012-12-03 11:13 ` Florian Westphal
2012-12-03 12:58 ` Pablo Neira Ayuso
2012-11-27 12:39 ` Pablo Neira Ayuso
2012-11-16 10:02 ` [PATCH 0/3] netfilter: add connlabel conntrack extension Pablo Neira Ayuso
2012-11-16 11:31 ` Florian Westphal [this message]
2012-11-16 11:52 ` 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=20121116113109.GK20678@breakpoint.cc \
--to=fw@strlen.de \
--cc=netfilter-devel@vger.kernel.org \
--cc=pablo@netfilter.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).