Merge branch 'Implement-classifier-action-terse-dump-mode'

Vlad Buslov says:

====================
Implement classifier-action terse dump mode

Output rate of current upstream kernel TC filter dump implementation if
relatively low (~100k rules/sec depending on configuration). This
constraint impacts performance of software switch implementation that
rely on TC for their datapath implementation and periodically call TC
filter dump to update rules stats. Moreover, TC filter dump output a lot
of static data that don't change during the filter lifecycle (filter
key, specific action details, etc.) which constitutes significant
portion of payload on resulting netlink packets and increases amount of
syscalls necessary to dump all filters on particular Qdisc. In order to
significantly improve filter dump rate this patch sets implement new
mode of TC filter dump operation named "terse dump" mode. In this mode
only parameters necessary to identify the filter (handle, action cookie,
etc.) and data that can change during filter lifecycle (filter flags,
action stats, etc.) are preserved in dump output while everything else
is omitted.

Userspace API is implemented using new TCA_DUMP_FLAGS tlv with only
available flag value TCA_DUMP_FLAGS_TERSE. Internally, new API requires
individual classifier support (new tcf_proto_ops->terse_dump()
callback). Support for action terse dump is implemented in act API and
don't require changing individual action implementations.

The following table provides performance comparison between regular
filter dump and new terse dump mode for two classifier-action profiles:
one minimal config with L2 flower classifier and single gact action and
another heavier config with L2+5tuple flower classifier with
tunnel_key+mirred actions.

 Classifier-action type      |        dump |  terse dump | X improvement
                             | (rules/sec) | (rules/sec) |
-----------------------------+-------------+-------------+---------------
 L2 with gact                |       141.8 |       293.2 |          2.07
 L2+5tuple tunnel_key+mirred |        76.4 |       198.8 |          2.60

Benchmark details: to measure the rate tc filter dump and terse dump
commands are invoked on ingress Qdisc that have one million filters
configured using following commands.

> time sudo tc -s filter show dev ens1f0 ingress >/dev/null

> time sudo tc -s filter show terse dev ens1f0 ingress >/dev/null

Value in results table is calculated by dividing 1000000 total rules by
"real" time reported by time command.

Setup details: 2x Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz, 32GB memory
====================

Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 58 insertions, 9 deletions

diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index 299b963c796e..752d608f4442 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -1851,7 +1851,7 @@ static int tcf_fill_node(struct net *net, struct sk_buff *skb,
 			 struct tcf_proto *tp, struct tcf_block *block,
 			 struct Qdisc *q, u32 parent, void *fh,
 			 u32 portid, u32 seq, u16 flags, int event,
-			 bool rtnl_held)
+			 bool terse_dump, bool rtnl_held)
 {
 	struct tcmsg *tcm;
 	struct nlmsghdr  *nlh;
@@ -1878,6 +1878,14 @@ static int tcf_fill_node(struct net *net, struct sk_buff *skb,
 		goto nla_put_failure;
 	if (!fh) {
 		tcm->tcm_handle = 0;
+	} else if (terse_dump) {
+		if (tp->ops->terse_dump) {
+			if (tp->ops->terse_dump(net, tp, fh, skb, tcm,
+						rtnl_held) < 0)
+				goto nla_put_failure;
+		} else {
+			goto cls_op_not_supp;
+		}
 	} else {
 		if (tp->ops->dump &&
 		    tp->ops->dump(net, tp, fh, skb, tcm, rtnl_held) < 0)
@@ -1888,6 +1896,7 @@ static int tcf_fill_node(struct net *net, struct sk_buff *skb,
 
 out_nlmsg_trim:
 nla_put_failure:
+cls_op_not_supp:
 	nlmsg_trim(skb, b);
 	return -1;
 }
@@ -1908,7 +1917,7 @@ static int tfilter_notify(struct net *net, struct sk_buff *oskb,
 
 	if (tcf_fill_node(net, skb, tp, block, q, parent, fh, portid,
 			  n->nlmsg_seq, n->nlmsg_flags, event,
-			  rtnl_held) <= 0) {
+			  false, rtnl_held) <= 0) {
 		kfree_skb(skb);
 		return -EINVAL;
 	}
@@ -1940,7 +1949,7 @@ static int tfilter_del_notify(struct net *net, struct sk_buff *oskb,
 
 	if (tcf_fill_node(net, skb, tp, block, q, parent, fh, portid,
 			  n->nlmsg_seq, n->nlmsg_flags, RTM_DELTFILTER,
-			  rtnl_held) <= 0) {
+			  false, rtnl_held) <= 0) {
 		NL_SET_ERR_MSG(extack, "Failed to build del event notification");
 		kfree_skb(skb);
 		return -EINVAL;
@@ -2501,6 +2510,7 @@ struct tcf_dump_args {
 	struct tcf_block *block;
 	struct Qdisc *q;
 	u32 parent;
+	bool terse_dump;
 };
 
 static int tcf_node_dump(struct tcf_proto *tp, void *n, struct tcf_walker *arg)
@@ -2511,12 +2521,12 @@ static int tcf_node_dump(struct tcf_proto *tp, void *n, struct tcf_walker *arg)
 	return tcf_fill_node(net, a->skb, tp, a->block, a->q, a->parent,
 			     n, NETLINK_CB(a->cb->skb).portid,
 			     a->cb->nlh->nlmsg_seq, NLM_F_MULTI,
-			     RTM_NEWTFILTER, true);
+			     RTM_NEWTFILTER, a->terse_dump, true);
 }
 
 static bool tcf_chain_dump(struct tcf_chain *chain, struct Qdisc *q, u32 parent,
 			   struct sk_buff *skb, struct netlink_callback *cb,
-			   long index_start, long *p_index)
+			   long index_start, long *p_index, bool terse)
 {
 	struct net *net = sock_net(skb->sk);
 	struct tcf_block *block = chain->block;
@@ -2545,7 +2555,7 @@ static bool tcf_chain_dump(struct tcf_chain *chain, struct Qdisc *q, u32 parent,
 			if (tcf_fill_node(net, skb, tp, block, q, parent, NULL,
 					  NETLINK_CB(cb->skb).portid,
 					  cb->nlh->nlmsg_seq, NLM_F_MULTI,
-					  RTM_NEWTFILTER, true) <= 0)
+					  RTM_NEWTFILTER, false, true) <= 0)
 				goto errout;
 			cb->args[1] = 1;
 		}
@@ -2561,6 +2571,7 @@ static bool tcf_chain_dump(struct tcf_chain *chain, struct Qdisc *q, u32 parent,
 		arg.w.skip = cb->args[1] - 1;
 		arg.w.count = 0;
 		arg.w.cookie = cb->args[2];
+		arg.terse_dump = terse;
 		tp->ops->walk(tp, &arg.w, true);
 		cb->args[2] = arg.w.cookie;
 		cb->args[1] = arg.w.count + 1;
@@ -2574,6 +2585,10 @@ errout:
 	return false;
 }
 
+static const struct nla_policy tcf_tfilter_dump_policy[TCA_MAX + 1] = {
+	[TCA_DUMP_FLAGS] = NLA_POLICY_BITFIELD32(TCA_DUMP_FLAGS_TERSE),
+};
+
 /* called with RTNL */
 static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb)
 {
@@ -2583,6 +2598,7 @@ static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb)
 	struct Qdisc *q = NULL;
 	struct tcf_block *block;
 	struct tcmsg *tcm = nlmsg_data(cb->nlh);
+	bool terse_dump = false;
 	long index_start;
 	long index;
 	u32 parent;
@@ -2592,10 +2608,17 @@ static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb)
 		return skb->len;
 
 	err = nlmsg_parse_deprecated(cb->nlh, sizeof(*tcm), tca, TCA_MAX,
-				     NULL, cb->extack);
+				     tcf_tfilter_dump_policy, cb->extack);
 	if (err)
 		return err;
 
+	if (tca[TCA_DUMP_FLAGS]) {
+		struct nla_bitfield32 flags =
+			nla_get_bitfield32(tca[TCA_DUMP_FLAGS]);
+
+		terse_dump = flags.value & TCA_DUMP_FLAGS_TERSE;
+	}
+
 	if (tcm->tcm_ifindex == TCM_IFINDEX_MAGIC_BLOCK) {
 		block = tcf_block_refcnt_get(net, tcm->tcm_block_index);
 		if (!block)
@@ -2653,7 +2676,7 @@ static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb)
 		    nla_get_u32(tca[TCA_CHAIN]) != chain->index)
 			continue;
 		if (!tcf_chain_dump(chain, q, parent, skb, cb,
-				    index_start, &index)) {
+				    index_start, &index, terse_dump)) {
 			tcf_chain_put(chain);
 			err = -EMSGSIZE;
 			break;
@@ -3156,7 +3179,8 @@ int tcf_exts_dump(struct sk_buff *skb, struct tcf_exts *exts)
 			if (nest == NULL)
 				goto nla_put_failure;
 
-			if (tcf_action_dump(skb, exts->actions, 0, 0) < 0)
+			if (tcf_action_dump(skb, exts->actions, 0, 0, false)
+			    < 0)
 				goto nla_put_failure;
 			nla_nest_end(skb, nest);
 		} else if (exts->police) {
@@ -3180,6 +3204,31 @@ nla_put_failure:
 }
 EXPORT_SYMBOL(tcf_exts_dump);
 
+int tcf_exts_terse_dump(struct sk_buff *skb, struct tcf_exts *exts)
+{
+#ifdef CONFIG_NET_CLS_ACT
+	struct nlattr *nest;
+
+	if (!exts->action || !tcf_exts_has_actions(exts))
+		return 0;
+
+	nest = nla_nest_start_noflag(skb, exts->action);
+	if (!nest)
+		goto nla_put_failure;
+
+	if (tcf_action_dump(skb, exts->actions, 0, 0, true) < 0)
+		goto nla_put_failure;
+	nla_nest_end(skb, nest);
+	return 0;
+
+nla_put_failure:
+	nla_nest_cancel(skb, nest);
+	return -1;
+#else
+	return 0;
+#endif
+}
+EXPORT_SYMBOL(tcf_exts_terse_dump);
 
 int tcf_exts_dump_stats(struct sk_buff *skb, struct tcf_exts *exts)
 {