Message-Id: <1382960120-2198-1-git-send-email-dborkman@redhat.com>
Date:	Mon, 28 Oct 2013 12:35:20 +0100
From:	Daniel Borkmann <dborkman@...hat.com>
To:	davem@...emloft.net
Cc:	netdev@...r.kernel.org, Thomas Graf <tgraf@...g.ch>
Subject: [PATCH net-next] net: sched: cls_bpf: add BPF-based classifier

This work contains a lightweight BPF-based traffic classifier that can
serve as a flexible alternative to ematch-based tree classification, i.e.
now that BPF filter engine can also be JITed in the kernel. Naturally, tc
actions and policies are supported as well with cls_bpf. Multiple BPF
programs/filter can be attached for a class, or they can just as well be
written within a single BPF program, that's really up to the user how he
wishes to run/optimize the code, e.g. also for inversion of verdicts etc.
The notion of a BPF program's return/exit codes is being kept as follows:
non-zero for match, zero for mismatch.

As a minimal usage example with iproute2, we use a 3 band prio root qdisc
on a router with sfq each as leave, and assign ssh and icmp bpf-based
filters to band 1, http traffic to band 2 and the rest to band 3. For the
first two bands we load the bytecode from a file, in the 2nd we load it
inline as an example:

echo 1 > /proc/sys/net/core/bpf_jit_enable

tc qdisc del dev em1 root
tc qdisc add dev em1 root handle 1: prio bands 3 priomap 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

tc qdisc add dev em1 parent 1:1 sfq perturb 16
tc qdisc add dev em1 parent 1:2 sfq perturb 16
tc qdisc add dev em1 parent 1:3 sfq perturb 16

tc filter add dev em1 parent 1: bpf run bytecode-file /etc/tc/ssh.bpf flowid 1:1
tc filter add dev em1 parent 1: bpf run bytecode-file /etc/tc/icmp.bpf flowid 1:1
tc filter add dev em1 parent 1: bpf run bytecode-file /etc/tc/http.bpf flowid 1:2
tc filter add dev em1 parent 1: bpf run bytecode "`bpfc -f tc -i misc.ops`" flowid 1:3

BPF programs can be easily created and passed to tc, either as inline
'bytecode' or 'bytecode-file'. There are a couple of front-ends that can
compile opcodes, for example:

1) People familiar with tcpdump-like filters:

   tcpdump -iem1 -ddd port 22 | tr '\n' ',' > /etc/tc/ssh.bpf

2) People that want to low-level program their filters or use BPF
   extensions that lack support by libpcap's compiler:

   bpfc -f tc -i ssh.ops > /etc/tc/ssh.bpf

   ssh.ops example code:
   ldh [12]
   jne #0x800, drop
   ldb [23]
   jneq #6, drop
   ldh [20]
   jset #0x1fff, drop
   ldxb 4 * ([14] & 0xf)
   ldh [%x + 14]
   jeq #0x16, pass
   ldh [%x + 16]
   jne #0x16, drop
   pass: ret #-1
   drop: ret #0

It was chosen to load bytecode into tc, since the reverse operation,
tc filter list dev em1, is then able to show the exact commands again.
Possible follow-up work could also include a small expression compiler
for iproute2. Tested with the help of bmon. This idea came up during
the Netfilter Workshop 2013 in Copenhagen.

Signed-off-by: Daniel Borkmann <dborkman@...hat.com>
Cc: Thomas Graf <tgraf@...g.ch>
---
 include/uapi/linux/pkt_cls.h |  14 ++
 net/sched/Kconfig            |  10 ++
 net/sched/Makefile           |   1 +
 net/sched/cls_bpf.c          | 380 +++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 405 insertions(+)
 create mode 100644 net/sched/cls_bpf.c

diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h
index 082eafa..25731df 100644
--- a/include/uapi/linux/pkt_cls.h
+++ b/include/uapi/linux/pkt_cls.h
@@ -388,6 +388,20 @@ enum {
 
 #define TCA_CGROUP_MAX (__TCA_CGROUP_MAX - 1)
 
+/* BPF classifier */
+
+enum {
+	TCA_BPF_UNSPEC,
+	TCA_BPF_ACT,
+	TCA_BPF_POLICE,
+	TCA_BPF_CLASSID,
+	TCA_BPF_OPS_LEN,
+	TCA_BPF_OPS,
+	__TCA_BPF_MAX,
+};
+
+#define TCA_BPF_MAX (__TCA_BPF_MAX - 1)
+
 /* Extended Matches */
 
 struct tcf_ematch_tree_hdr {
diff --git a/net/sched/Kconfig b/net/sched/Kconfig
index c03a32a..989464c 100644
--- a/net/sched/Kconfig
+++ b/net/sched/Kconfig
@@ -443,6 +443,16 @@ config NET_CLS_CGROUP
 	  To compile this code as a module, choose M here: the
 	  module will be called cls_cgroup.
 
+config NET_CLS_BPF
+	tristate "BPF-based classifier"
+	select NET_CLS
+	---help---
+	  If you say Y here, you will be able to classify packets based on
+	  programmable BPF (JIT'ed) filters as an alternative to ematches.
+
+	  To compile this code as a module, choose M here: the module will
+	  be called cls_bpf.
+
 config NET_EMATCH
 	bool "Extended Matches"
 	select NET_CLS
diff --git a/net/sched/Makefile b/net/sched/Makefile
index e5f9abe..35fa47a 100644
--- a/net/sched/Makefile
+++ b/net/sched/Makefile
@@ -50,6 +50,7 @@ obj-$(CONFIG_NET_CLS_RSVP6)	+= cls_rsvp6.o
 obj-$(CONFIG_NET_CLS_BASIC)	+= cls_basic.o
 obj-$(CONFIG_NET_CLS_FLOW)	+= cls_flow.o
 obj-$(CONFIG_NET_CLS_CGROUP)	+= cls_cgroup.o
+obj-$(CONFIG_NET_CLS_BPF)	+= cls_bpf.o
 obj-$(CONFIG_NET_EMATCH)	+= ematch.o
 obj-$(CONFIG_NET_EMATCH_CMP)	+= em_cmp.o
 obj-$(CONFIG_NET_EMATCH_NBYTE)	+= em_nbyte.o
diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c
new file mode 100644
index 0000000..778ca86
--- /dev/null
+++ b/net/sched/cls_bpf.c
@@ -0,0 +1,380 @@
+/*
+ * Berkeley Packet Filter based traffic classifier
+ *
+ * Might be used to classify traffic through flexible, user-defined and
+ * possibly JIT-ed BPF filters for traffic control as an alternative to
+ * ematches.
+ *
+ * (C) 2013 Daniel Borkmann <dborkman@...hat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/skbuff.h>
+#include <linux/filter.h>
+#include <net/rtnetlink.h>
+#include <net/pkt_cls.h>
+#include <net/sock.h>
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Daniel Borkmann <dborkman@...hat.com>");
+MODULE_DESCRIPTION("TC BPF based classifier");
+
+struct cls_bpf_head {
+	struct list_head plist;
+	u32 hgen;
+};
+
+struct cls_bpf_prog {
+	struct sk_filter *filter;
+	struct sock_filter *bpf_ops;
+	struct tcf_exts exts;
+	struct tcf_result res;
+	struct list_head link;
+	u32 handle;
+	u16 bpf_len;
+};
+
+static const struct nla_policy bpf_policy[TCA_BPF_MAX + 1] = {
+	[TCA_BPF_CLASSID]	= { .type = NLA_U32 },
+	[TCA_BPF_OPS_LEN]	= { .type = NLA_U16 },
+	[TCA_BPF_OPS]		= { .type = NLA_BINARY,
+				    .len = sizeof(struct sock_filter) * BPF_MAXINSNS },
+};
+
+static const struct tcf_ext_map bpf_ext_map = {
+	.action = TCA_BPF_ACT,
+	.police = TCA_BPF_POLICE,
+};
+
+static int cls_bpf_classify(struct sk_buff *skb, const struct tcf_proto *tp,
+			    struct tcf_result *res)
+{
+	struct cls_bpf_head *head = tp->root;
+	struct cls_bpf_prog *prog;
+	int ret;
+
+	list_for_each_entry(prog, &head->plist, link) {
+		if (SK_RUN_FILTER(prog->filter, skb) == 0)
+			continue;
+
+		*res = prog->res;
+		ret = tcf_exts_exec(skb, &prog->exts, res);
+		if (ret < 0)
+			continue;
+
+		return ret;
+	}
+
+	return -1;
+}
+
+static int cls_bpf_init(struct tcf_proto *tp)
+{
+	struct cls_bpf_head *head;
+
+	head = kzalloc(sizeof(*head), GFP_KERNEL);
+	if (head == NULL)
+		return -ENOBUFS;
+
+	INIT_LIST_HEAD(&head->plist);
+	tp->root = head;
+
+	return 0;
+}
+
+static void cls_bpf_delete_prog(struct tcf_proto *tp, struct cls_bpf_prog *prog)
+{
+	tcf_unbind_filter(tp, &prog->res);
+	tcf_exts_destroy(tp, &prog->exts);
+
+	sk_unattached_filter_destroy(prog->filter);
+
+	kfree(prog->bpf_ops);
+	kfree(prog);
+}
+
+static int cls_bpf_delete(struct tcf_proto *tp, unsigned long arg)
+{
+	struct cls_bpf_head *head = tp->root;
+	struct cls_bpf_prog *prog, *todel = (struct cls_bpf_prog *) arg;
+
+	list_for_each_entry(prog, &head->plist, link) {
+		if (prog == todel) {
+			tcf_tree_lock(tp);
+			list_del(&prog->link);
+			tcf_tree_unlock(tp);
+
+			cls_bpf_delete_prog(tp, prog);
+			return 0;
+		}
+	}
+
+	return -ENOENT;
+}
+
+static void cls_bpf_destroy(struct tcf_proto *tp)
+{
+	struct cls_bpf_head *head = tp->root;
+	struct cls_bpf_prog *prog, *tmp;
+
+	list_for_each_entry_safe(prog, tmp, &head->plist, link) {
+		list_del(&prog->link);
+		cls_bpf_delete_prog(tp, prog);
+	}
+
+	kfree(head);
+}
+
+static unsigned long cls_bpf_get(struct tcf_proto *tp, u32 handle)
+{
+	struct cls_bpf_head *head = tp->root;
+	struct cls_bpf_prog *prog;
+	unsigned long ret = 0UL;
+
+	if (head == NULL)
+		return 0UL;
+
+	list_for_each_entry(prog, &head->plist, link) {
+		if (prog->handle == handle) {
+			ret = (unsigned long) prog;
+			break;
+		}
+	}
+
+	return ret;
+}
+
+static void cls_bpf_put(struct tcf_proto *tp, unsigned long f)
+{
+}
+
+static int cls_bpf_modify_existing(struct net *net, struct tcf_proto *tp,
+				   struct cls_bpf_prog *prog,
+				   unsigned long base, struct nlattr **tb,
+				   struct nlattr *est)
+{
+	struct sock_filter *bpf_ops, *bpf_old;
+	struct tcf_exts exts;
+	struct sock_fprog tmp;
+	struct sk_filter *fp, *fp_old;
+	u16 bpf_size, bpf_len;
+	u32 classid;
+	int ret;
+
+	if (!tb[TCA_BPF_OPS_LEN] || !tb[TCA_BPF_OPS] || !tb[TCA_BPF_CLASSID])
+		return -EINVAL;
+
+	ret = tcf_exts_validate(net, tp, tb, est, &exts, &bpf_ext_map);
+	if (ret < 0)
+		return ret;
+
+	classid = nla_get_u32(tb[TCA_BPF_CLASSID]);
+	bpf_len = nla_get_u16(tb[TCA_BPF_OPS_LEN]);
+	if (bpf_len > BPF_MAXINSNS || bpf_len == 0) {
+		ret = -EINVAL;
+		goto errout;
+	}
+
+	bpf_size = bpf_len * sizeof(*bpf_ops);
+	bpf_ops = kzalloc(bpf_size, GFP_KERNEL);
+	if (bpf_ops == NULL) {
+		ret = -ENOMEM;
+		goto errout;
+	}
+
+	memcpy(bpf_ops, nla_data(tb[TCA_BPF_OPS]), bpf_size);
+
+	tmp.len = bpf_len;
+	tmp.filter = (struct sock_filter __user *) bpf_ops;
+
+	ret = sk_unattached_filter_create(&fp, &tmp);
+	if (ret)
+		goto errout_free;
+
+	tcf_tree_lock(tp);
+	fp_old = prog->filter;
+	bpf_old = prog->bpf_ops;
+
+	prog->bpf_len = bpf_len;
+	prog->bpf_ops = bpf_ops;
+	prog->filter = fp;
+	prog->res.classid = classid;
+	tcf_tree_unlock(tp);
+
+	tcf_bind_filter(tp, &prog->res, base);
+	tcf_exts_change(tp, &prog->exts, &exts);
+
+	if (fp_old)
+		sk_unattached_filter_destroy(fp_old);
+	if (bpf_old)
+		kfree(bpf_old);
+
+	return 0;
+
+errout_free:
+	kfree(bpf_ops);
+errout:
+	tcf_exts_destroy(tp, &exts);
+	return ret;
+}
+
+static u32 cls_bpf_grab_new_handle(struct tcf_proto *tp,
+				   struct cls_bpf_head *head)
+{
+	unsigned int i = 0x80000000;
+
+	do {
+		if (++head->hgen == 0x7FFFFFFF)
+			head->hgen = 1;
+	} while (--i > 0 && cls_bpf_get(tp, head->hgen));
+	if (i == 0)
+		pr_err("Insufficient number of handles\n");
+
+	return i;
+}
+
+static int cls_bpf_change(struct net *net, struct sk_buff *in_skb,
+			  struct tcf_proto *tp, unsigned long base,
+			  u32 handle, struct nlattr **tca,
+			  unsigned long *arg)
+{
+	struct cls_bpf_head *head = tp->root;
+	struct cls_bpf_prog *prog = (struct cls_bpf_prog *) *arg;
+	struct nlattr *tb[TCA_BPF_MAX + 1];
+	int ret;
+
+	if (tca[TCA_OPTIONS] == NULL)
+		return -EINVAL;
+
+	ret = nla_parse_nested(tb, TCA_BPF_MAX, tca[TCA_OPTIONS], bpf_policy);
+	if (ret < 0)
+		return ret;
+
+	if (prog != NULL) {
+		if (handle && prog->handle != handle)
+			return -EINVAL;
+		return cls_bpf_modify_existing(net, tp, prog, base, tb,
+					       tca[TCA_RATE]);
+	}
+
+	prog = kzalloc(sizeof(*prog), GFP_KERNEL);
+	if (prog == NULL)
+		return -ENOBUFS;
+
+	if (handle == 0)
+		prog->handle = cls_bpf_grab_new_handle(tp, head);
+	else
+		prog->handle = handle;
+	if (prog->handle == 0) {
+		ret = -EINVAL;
+		goto errout;
+	}
+
+	ret = cls_bpf_modify_existing(net, tp, prog, base, tb, tca[TCA_RATE]);
+	if (ret < 0)
+		goto errout;
+
+	tcf_tree_lock(tp);
+	list_add(&prog->link, &head->plist);
+	tcf_tree_unlock(tp);
+
+	*arg = (unsigned long) prog;
+
+	return 0;
+errout:
+	if (*arg == 0UL && prog)
+		kfree(prog);
+
+	return ret;
+}
+
+static int cls_bpf_dump(struct tcf_proto *tp, unsigned long fh,
+			struct sk_buff *skb, struct tcmsg *tm)
+{
+	struct cls_bpf_prog *prog = (struct cls_bpf_prog *) fh;
+	struct nlattr *nest, *nla;
+
+	if (prog == NULL)
+		return skb->len;
+
+	tm->tcm_handle = prog->handle;
+
+	nest = nla_nest_start(skb, TCA_OPTIONS);
+	if (nest == NULL)
+		goto nla_put_failure;
+
+	if (nla_put_u32(skb, TCA_BPF_CLASSID, prog->res.classid))
+		goto nla_put_failure;
+	if (nla_put_u16(skb, TCA_BPF_OPS_LEN, prog->bpf_len))
+		goto nla_put_failure;
+
+	nla = nla_reserve(skb, TCA_BPF_OPS, prog->bpf_len *
+			  sizeof(struct sock_filter));
+	if (nla == NULL)
+		goto nla_put_failure;
+
+        memcpy(nla_data(nla), prog->bpf_ops, nla_len(nla));
+
+	if (tcf_exts_dump(skb, &prog->exts, &bpf_ext_map) < 0)
+		goto nla_put_failure;
+
+	nla_nest_end(skb, nest);
+
+	if (tcf_exts_dump_stats(skb, &prog->exts, &bpf_ext_map) < 0)
+		goto nla_put_failure;
+
+	return skb->len;
+
+nla_put_failure:
+	nla_nest_cancel(skb, nest);
+	return -1;
+}
+
+static void cls_bpf_walk(struct tcf_proto *tp, struct tcf_walker *arg)
+{
+	struct cls_bpf_head *head = tp->root;
+	struct cls_bpf_prog *prog;
+
+	list_for_each_entry(prog, &head->plist, link) {
+		if (arg->count < arg->skip)
+			goto skip;
+		if (arg->fn(tp, (unsigned long) prog, arg) < 0) {
+			arg->stop = 1;
+			break;
+		}
+skip:
+		arg->count++;
+	}
+}
+
+static struct tcf_proto_ops cls_bpf_ops __read_mostly = {
+	.kind		=	"bpf",
+	.owner		=	THIS_MODULE,
+	.classify	=	cls_bpf_classify,
+	.init		=	cls_bpf_init,
+	.destroy	=	cls_bpf_destroy,
+	.get		=	cls_bpf_get,
+	.put		=	cls_bpf_put,
+	.change		=	cls_bpf_change,
+	.delete		=	cls_bpf_delete,
+	.walk		=	cls_bpf_walk,
+	.dump		=	cls_bpf_dump,
+};
+
+static int __init cls_bpf_init_mod(void)
+{
+	return register_tcf_proto_ops(&cls_bpf_ops);
+}
+
+static void __exit cls_bpf_exit_mod(void)
+{
+	unregister_tcf_proto_ops(&cls_bpf_ops);
+}
+
+module_init(cls_bpf_init_mod);
+module_exit(cls_bpf_exit_mod);
-- 
1.8.3.1

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

Powered by blists - more mailing lists