diff mbox

[lnf-queue] src: PF_BIND/UNBIND is ignored in 3.8 and later

Message ID 1398330071-3463-1-git-send-email-fw@strlen.de
State Accepted
Headers show

Commit Message

Florian Westphal April 24, 2014, 9:01 a.m. UTC
There is confusion on what this command actually does and why
examples commonly PF_UNBIND at startup.

Since these are obsolete document that its not needed at all
from 3.8 onwards and remove the unbind from the example program.

Signed-off-by: Florian Westphal <fw@strlen.de>
---
I'll push it later today if noone objets.

Comments

Pablo Neira Ayuso April 24, 2014, 10:01 a.m. UTC | #1
On Thu, Apr 24, 2014 at 11:01:11AM +0200, Florian Westphal wrote:
> There is confusion on what this command actually does and why
> examples commonly PF_UNBIND at startup.
> 
> Since these are obsolete document that its not needed at all
> from 3.8 onwards and remove the unbind from the example program.
> 
> Signed-off-by: Florian Westphal <fw@strlen.de>
> ---
> I'll push it later today if noone objets.
> 
> diff --git a/examples/nf-queue.c b/examples/nf-queue.c
> index 1f465ad..4e0b227 100644
> --- a/examples/nf-queue.c
> +++ b/examples/nf-queue.c
> @@ -136,17 +136,9 @@ int main(int argc, char *argv[])
>  		exit(EXIT_FAILURE);
>  	}
>  
> -	nlh = nfq_hdr_put(buf, NFQNL_MSG_CONFIG, 0);
> -	nfq_nlmsg_cfg_put_cmd(nlh, AF_INET, NFQNL_CFG_CMD_PF_UNBIND);
> -
> -	if (mnl_socket_sendto(nl, nlh, nlh->nlmsg_len) < 0) {
> -		perror("mnl_socket_send");
> -		exit(EXIT_FAILURE);
> -	}
> -
> +	/* PF_BIND is not needed with kernels 3.8 and later */

My only concern is that people using old kernels won't manage to run
this example, I think there are quite a lot of people running < 3.8,
so we may hit confusion from the other side.

I like the remaining part of the patch, but this part, I would just
leave it with a big fat comment like from here to there you can remove
this code since the NFQNL_CFG_CMD_PF_UNBIND is noop in >= 3.8.
--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Florian Westphal April 24, 2014, 10:09 a.m. UTC | #2
Pablo Neira Ayuso <pablo@netfilter.org> wrote:
> > -	nlh = nfq_hdr_put(buf, NFQNL_MSG_CONFIG, 0);
> > -	nfq_nlmsg_cfg_put_cmd(nlh, AF_INET, NFQNL_CFG_CMD_PF_UNBIND);
> > -
> > -	if (mnl_socket_sendto(nl, nlh, nlh->nlmsg_len) < 0) {
> > -		perror("mnl_socket_send");
> > -		exit(EXIT_FAILURE);
> > -	}
> > -
> > +	/* PF_BIND is not needed with kernels 3.8 and later */
> 
> My only concern is that people using old kernels won't manage to run
> this example, I think there are quite a lot of people running < 3.8,
> so we may hit confusion from the other side.

I pushed the patch without the UNBIND removal.

Thanks.
--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/examples/nf-queue.c b/examples/nf-queue.c
index 1f465ad..4e0b227 100644
--- a/examples/nf-queue.c
+++ b/examples/nf-queue.c
@@ -136,17 +136,9 @@  int main(int argc, char *argv[])
 		exit(EXIT_FAILURE);
 	}
 
-	nlh = nfq_hdr_put(buf, NFQNL_MSG_CONFIG, 0);
-	nfq_nlmsg_cfg_put_cmd(nlh, AF_INET, NFQNL_CFG_CMD_PF_UNBIND);
-
-	if (mnl_socket_sendto(nl, nlh, nlh->nlmsg_len) < 0) {
-		perror("mnl_socket_send");
-		exit(EXIT_FAILURE);
-	}
-
+	/* PF_BIND is not needed with kernels 3.8 and later */
 	nlh = nfq_hdr_put(buf, NFQNL_MSG_CONFIG, 0);
 	nfq_nlmsg_cfg_put_cmd(nlh, AF_INET, NFQNL_CFG_CMD_PF_BIND);
-
 	if (mnl_socket_sendto(nl, nlh, nlh->nlmsg_len) < 0) {
 		perror("mnl_socket_send");
 		exit(EXIT_FAILURE);
diff --git a/src/libnetfilter_queue.c b/src/libnetfilter_queue.c
index 32725d1..740b340 100644
--- a/src/libnetfilter_queue.c
+++ b/src/libnetfilter_queue.c
@@ -455,8 +455,9 @@  EXPORT_SYMBOL(nfq_close);
  * \param h Netfilter queue connection handle obtained via call to nfq_open()
  * \param pf protocol family to bind to nfqueue handler obtained from nfq_open()
  *
- * Binds the given queue connection handle to process packets belonging to 
+ * Binds the given queue connection handle to process packets belonging to
  * the given protocol family (ie. PF_INET, PF_INET6, etc).
+ * This call is obsolete, Linux kernels from 3.8 onwards ignore it.
  *
  * \return integer inferior to 0 in case of failure
  */
@@ -473,6 +474,8 @@  EXPORT_SYMBOL(nfq_bind_pf);
  *
  * Unbinds the given queue connection handle from processing packets belonging
  * to the given protocol family.
+ *
+ * This call is obsolete, Linux kernels from 3.8 onwards ignore it.
  */
 int nfq_unbind_pf(struct nfq_handle *h, u_int16_t pf)
 {
diff --git a/src/nlmsg.c b/src/nlmsg.c
index 81e170e..aebdd5e 100644
--- a/src/nlmsg.c
+++ b/src/nlmsg.c
@@ -77,10 +77,13 @@  EXPORT_SYMBOL(nfq_nlmsg_verdict_put_pkt);
  *   subsystem is working.
  * - NFQNL_CFG_CMD_BIND: Binds the program to a specific queue.
  * - NFQNL_CFG_CMD_UNBIND: Unbinds the program to a specifiq queue.
+ *
+ * Obsolete commands:
  * - NFQNL_CFG_CMD_PF_BIND: Binds to process packets belonging to the given
  *   protocol family (ie. PF_INET, PF_INET6, etc).
  * - NFQNL_CFG_CMD_PF_UNBIND: Unbinds from processing packets belonging to the
- *   given protocol family.
+ *   given protocol family.  Both commands are ignored by Linux kernel 3.8 and
+ *   later versions.
  */
 void nfq_nlmsg_cfg_put_cmd(struct nlmsghdr *nlh, uint16_t pf, uint8_t cmd)
 {