All of lore.kernel.org
 help / color / mirror / Atom feed
From: Duncan Roe <duncan_roe@optusnet.com.au>
To: Florian Westphal <fw@strlen.de>, Pablo Neira Ayuso <pablo@netfilter.org>
Cc: Netfilter Development <netfilter-devel@vger.kernel.org>
Subject: Re: [PATCH libnetfilter_queue v3 1-5/5] src: Speed-up
Date: Mon, 7 Feb 2022 11:24:15 +1100	[thread overview]
Message-ID: <YgBmr3+IqGXQ4M6E@slk1.local.net> (raw)
In-Reply-To: <20220120120458.GF31905@breakpoint.cc>

On Thu, Jan 20, 2022 at 01:04:58PM +0100, Florian Westphal wrote:
> Duncan Roe <duncan_roe@optusnet.com.au> wrote:
> > On Tue, Jan 18, 2022 at 02:11:43AM +0100, Pablo Neira Ayuso wrote:
> > >
> > > This patch have a number of showstoppers such as exposing structure
> > > layout on the header files.
> > >
> > That's only in patch 5. You could apply 1-4. There are actually no other
> > showstoppers, right?
>
> Regarding patch 5, I think its ok except the pkt_buff layout freeze.
>
> From a quick glance, there is no assumption that the data area resides
> after the pktbuff head, so it should be possible to keep pkt_buff
> private, allocate an empty packet and then associate a new buffer with
> it.
>
> I agree the memcpy needs to go, nfqueue uses should use F_GSO feature
> flag and memcpy'ing 60k big packets isn't ideal.

There is no pkt_buff layout freeze. If we want to change it in future, we bump
the major version of libnetfilter_queue.so, same as we would do if changing the
signature of an existing function.

Or am I missing something?

Cheers ... Duncan.

  parent reply	other threads:[~2022-02-07  0:24 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-09  3:16 [PATCH libnetfilter_queue v3 0/5] Speed-up Duncan Roe
2022-01-09  3:16 ` [PATCH libnetfilter_queue v3 1/5] src: eliminate packet copy when constructing struct pkt_buff Duncan Roe
2022-01-09  3:16 ` [PATCH libnetfilter_queue v3 2/5] src: Avoid compiler warning Duncan Roe
2022-01-09  3:16 ` [PATCH libnetfilter_queue v3 3/5] src: Use more meaningful name in callback.c Duncan Roe
2022-01-09  3:16 ` [PATCH libnetfilter_queue v3 4/5] build: doc: Eliminate doxygen warnings Duncan Roe
2022-01-09  3:16 ` [PATCH libnetfilter_queue v3 5/5] src: struct pktbuff is no longer opaque Duncan Roe
2022-01-17 23:58   ` [PATCH libnetfilter_queue v3 1-5/5] src: Speed-up Duncan Roe
2022-01-18  1:11     ` Pablo Neira Ayuso
2022-01-20  3:56       ` Duncan Roe
2022-01-20  6:27         ` Florian Westphal
2022-01-21  2:36           ` Duncan Roe
2022-01-23  2:03             ` Duncan Roe
2022-01-20 12:04         ` Florian Westphal
2022-01-20 12:40           ` Pablo Neira Ayuso
2022-02-22  1:39             ` Duncan Roe
2022-02-07  0:24           ` Duncan Roe [this message]
2022-02-07 10:44             ` Florian Westphal
2022-01-21  3:19       ` Duncan Roe

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=YgBmr3+IqGXQ4M6E@slk1.local.net \
    --to=duncan_roe@optusnet.com.au \
    --cc=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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.