netfilter-devel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Pablo Neira Ayuso <pablo@netfilter.org>
To: Elise Lennion <elise.lennion@gmail.com>
Cc: netfilter-devel@vger.kernel.org
Subject: Re: [PATCH nft] datatype: Replace getnameinfo() by internal lookup table
Date: Thu, 24 Nov 2016 10:31:15 +0100	[thread overview]
Message-ID: <20161124093115.GA1461@salvia> (raw)
In-Reply-To: <20161123225317.GA20966@lennorien.com>

Hi Elise,

On Wed, Nov 23, 2016 at 08:53:17PM -0200, Elise Lennion wrote:
> To avoid exceeding the inputs number limit of the flex scanner used.
> 
> Using port number as index, to map service names in the table, results
> in a very sparse table, so a new struct is needed to associate ports
> with names.

Please, indicate what /etc/services you have used, ie. your linux
distro version, so that info remains here for the record.

More comments below.

> Signed-off-by: Elise Lennion <elise.lennion@gmail.com>
> ---
> 
>  The services list are all well-known and registered ports of my local
>  /etc/services file.
> 
>  include/datatype.h |  11 ++++
>  src/datatype.c     | 159 ++++++++++++++++++++++++++++++++++++++++++++++++++---
>  2 files changed, 162 insertions(+), 8 deletions(-)
> 
> diff --git a/include/datatype.h b/include/datatype.h
> index 9f3f711..9c34c50 100644
> --- a/include/datatype.h
> +++ b/include/datatype.h
> @@ -172,6 +172,17 @@ struct symbolic_constant {
>  	uint64_t			value;
>  };
>  
> +/**
> + * struct port_servname - port <-> service name mapping
> + *
> + * @port: port number
> + * @name: service name
> + */
> +struct port_servname {
> +	uint16_t			port;
> +	char				*name;
> +};
> +
>  #define SYMBOL(id, v)	{ .identifier = (id), .value = (v) }
>  #define SYMBOL_LIST_END	(struct symbolic_constant) { }
>  
> diff --git a/src/datatype.c b/src/datatype.c
> index 1e40287..85b261d 100644
> --- a/src/datatype.c
> +++ b/src/datatype.c
> @@ -28,6 +28,156 @@
>  
>  #include <netinet/ip_icmp.h>
>  
> +static const struct port_servname services[] = {

Given that this is long, please, place this and nft_service_lookup()
under a new file, src/services.c.

> +	{1, "tcpmux"}, {7, "echo"}, {9, "discard"},

We use symbol_table for these definitions instead, eg.

        static const struct symbol_table icmp_code_tbl = {
                ...
                .sym_tbl        = &icmp_code_tbl,

If you attach the symbol table that defines the services to the
inet_service_type definition, then "nft describe tcp dport" will also
display the available service names.

BTW, on a different front, the binary search in this patch is useful,
you can probably integrate this into the core, I mean
symbolic_constant_print(). Only problem is that we would need an
explicit size in struct symbol_table, but that shouldn't be a problem.
And update many spots in the code, but that should be fine. You can
follow up with a patch that applies on top of this to add this binary
search approach.

Thanks.

      reply	other threads:[~2016-11-24  9:31 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-11-23 22:53 [PATCH nft] datatype: Replace getnameinfo() by internal lookup table Elise Lennion
2016-11-24  9:31 ` 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=20161124093115.GA1461@salvia \
    --to=pablo@netfilter.org \
    --cc=elise.lennion@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).