diff mbox

[8/8] netfilter: implement xt_cgroup cgroup2 path match

Message ID 1449527935-27056-9-git-send-email-tj@kernel.org
State Accepted
Delegated to: Pablo Neira
Headers show

Commit Message

Tejun Heo Dec. 7, 2015, 10:38 p.m. UTC
This patch implements xt_cgroup path match which matches cgroup2
membership of the associated socket.  The match is recursive and
invertible.

For rationales on introducing another cgroup based match, please refer
to a preceding commit "sock, cgroup: add sock->sk_cgroup".

v3: Folded into xt_cgroup as a new revision interface as suggested by
    Pablo.

v2: Included linux/limits.h from xt_cgroup2.h for PATH_MAX.  Added
    explicit alignment to the priv field.  Both suggested by Jan.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Daniel Wagner <daniel.wagner@bmw-carit.de>
CC: Neil Horman <nhorman@tuxdriver.com>
Cc: Jan Engelhardt <jengelh@inai.de>
Cc: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/uapi/linux/netfilter/xt_cgroup.h | 13 ++++++
 net/netfilter/xt_cgroup.c                | 69 ++++++++++++++++++++++++++++++++
 2 files changed, 82 insertions(+)

Comments

Pablo Neira Ayuso Dec. 14, 2015, 7:37 p.m. UTC | #1
On Mon, Dec 07, 2015 at 05:38:55PM -0500, Tejun Heo wrote:
> This patch implements xt_cgroup path match which matches cgroup2
> membership of the associated socket.  The match is recursive and
> invertible.

Applied, thanks.

I shared the same concerns as Florian regarding the large size of the
path field in iptables, but given that we expose the layout of our
internal representation there (which is bad in terms of
extensibility), the only solution that I can see is to artificially
limitate the size of that field, but that may break users depending on
the scenario.

Hopefully, we should be able to provide something better in nf_tables
to address this.
--
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
Alban Crequy Feb. 12, 2016, 12:10 a.m. UTC | #2
Hi,

On 7 December 2015 at 23:38, Tejun Heo <tj@kernel.org> wrote:
> This patch implements xt_cgroup path match which matches cgroup2
> membership of the associated socket.  The match is recursive and
> invertible.

Is there any plans to implement a similar cgroup2 path match in a
cgroup classifier in tc?
I wonder if it will be possible to use cgroup to classify packets in
tc without net_cls.class_id in cgroup2.

Thanks!
Alban
--
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
Tejun Heo Feb. 12, 2016, 4:05 p.m. UTC | #3
Hello,

On Fri, Feb 12, 2016 at 01:10:18AM +0100, Alban Crequy wrote:
> Is there any plans to implement a similar cgroup2 path match in a
> cgroup classifier in tc?

That'd be great.  I wanted to but couldn't figure out the heads and
tails after staring at the code for half an hour.

> I wonder if it will be possible to use cgroup to classify packets in
> tc without net_cls.class_id in cgroup2.

You can still match against cgroup v2 path using xt_cgroup and then
mark it and then match that from tc, which isn't ideal but works.

Thanks.
diff mbox

Patch

diff --git a/include/uapi/linux/netfilter/xt_cgroup.h b/include/uapi/linux/netfilter/xt_cgroup.h
index 577c9e0..1e4b37b 100644
--- a/include/uapi/linux/netfilter/xt_cgroup.h
+++ b/include/uapi/linux/netfilter/xt_cgroup.h
@@ -2,10 +2,23 @@ 
 #define _UAPI_XT_CGROUP_H
 
 #include <linux/types.h>
+#include <linux/limits.h>
 
 struct xt_cgroup_info_v0 {
 	__u32 id;
 	__u32 invert;
 };
 
+struct xt_cgroup_info_v1 {
+	__u8		has_path;
+	__u8		has_classid;
+	__u8		invert_path;
+	__u8		invert_classid;
+	char		path[PATH_MAX];
+	__u32		classid;
+
+	/* kernel internal data */
+	void		*priv __attribute__((aligned(8)));
+};
+
 #endif /* _UAPI_XT_CGROUP_H */
diff --git a/net/netfilter/xt_cgroup.c b/net/netfilter/xt_cgroup.c
index 1730025..a086a91 100644
--- a/net/netfilter/xt_cgroup.c
+++ b/net/netfilter/xt_cgroup.c
@@ -34,6 +34,37 @@  static int cgroup_mt_check_v0(const struct xt_mtchk_param *par)
 	return 0;
 }
 
+static int cgroup_mt_check_v1(const struct xt_mtchk_param *par)
+{
+	struct xt_cgroup_info_v1 *info = par->matchinfo;
+	struct cgroup *cgrp;
+
+	if ((info->invert_path & ~1) || (info->invert_classid & ~1))
+		return -EINVAL;
+
+	if (!info->has_path && !info->has_classid) {
+		pr_info("xt_cgroup: no path or classid specified\n");
+		return -EINVAL;
+	}
+
+	if (info->has_path && info->has_classid) {
+		pr_info("xt_cgroup: both path and classid specified\n");
+		return -EINVAL;
+	}
+
+	if (info->has_path) {
+		cgrp = cgroup_get_from_path(info->path);
+		if (IS_ERR(cgrp)) {
+			pr_info("xt_cgroup: invalid path, errno=%ld\n",
+				PTR_ERR(cgrp));
+			return -EINVAL;
+		}
+		info->priv = cgrp;
+	}
+
+	return 0;
+}
+
 static bool
 cgroup_mt_v0(const struct sk_buff *skb, struct xt_action_param *par)
 {
@@ -46,6 +77,31 @@  cgroup_mt_v0(const struct sk_buff *skb, struct xt_action_param *par)
 		info->invert;
 }
 
+static bool cgroup_mt_v1(const struct sk_buff *skb, struct xt_action_param *par)
+{
+	const struct xt_cgroup_info_v1 *info = par->matchinfo;
+	struct sock_cgroup_data *skcd = &skb->sk->sk_cgrp_data;
+	struct cgroup *ancestor = info->priv;
+
+	if (!skb->sk || !sk_fullsock(skb->sk))
+		return false;
+
+	if (ancestor)
+		return cgroup_is_descendant(sock_cgroup_ptr(skcd), ancestor) ^
+			info->invert_path;
+	else
+		return (info->classid == sock_cgroup_classid(skcd)) ^
+			info->invert_classid;
+}
+
+static void cgroup_mt_destroy_v1(const struct xt_mtdtor_param *par)
+{
+	struct xt_cgroup_info_v1 *info = par->matchinfo;
+
+	if (info->priv)
+		cgroup_put(info->priv);
+}
+
 static struct xt_match cgroup_mt_reg[] __read_mostly = {
 	{
 		.name		= "cgroup",
@@ -59,6 +115,19 @@  static struct xt_match cgroup_mt_reg[] __read_mostly = {
 				  (1 << NF_INET_POST_ROUTING) |
 				  (1 << NF_INET_LOCAL_IN),
 	},
+	{
+		.name		= "cgroup",
+		.revision	= 1,
+		.family		= NFPROTO_UNSPEC,
+		.checkentry	= cgroup_mt_check_v1,
+		.match		= cgroup_mt_v1,
+		.matchsize	= sizeof(struct xt_cgroup_info_v1),
+		.destroy	= cgroup_mt_destroy_v1,
+		.me		= THIS_MODULE,
+		.hooks		= (1 << NF_INET_LOCAL_OUT) |
+				  (1 << NF_INET_POST_ROUTING) |
+				  (1 << NF_INET_LOCAL_IN),
+	},
 };
 
 static int __init cgroup_mt_init(void)