netfilter: nf_flow_table: count pending offload workqueue tasks

To improve hardware offload debuggability count pending 'add', 'del' and
'stats' flow_table offload workqueue tasks. Counters are incremented before
scheduling new task and decremented when workqueue handler finishes
executing. These counters allow user to diagnose congestion on hardware
offload workqueues that can happen when either CPU is starved and workqueue
jobs are executed at lower rate than new ones are added or when
hardware/driver can't keep up with the rate.

Implement the described counters as percpu counters inside new struct
netns_ft which is stored inside struct net. Expose them via new procfs file
'/proc/net/stats/nf_flowtable' that is similar to existing 'nf_conntrack'
file.

Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Signed-off-by: Oz Shlomo <ozsh@nvidia.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

5 files changed, 165 insertions, 4 deletions

diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index ddc54b6d18ee..df6abbfe0079 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -734,6 +734,15 @@ config NF_FLOW_TABLE
 
 	  To compile it as a module, choose M here.
 
+config NF_FLOW_TABLE_PROCFS
+	bool "Supply flow table statistics in procfs"
+	default y
+	depends on PROC_FS
+	depends on SYSCTL
+	help
+	  This option enables for the flow table offload statistics
+	  to be shown in procfs under net/netfilter/nf_flowtable.
+
 config NETFILTER_XTABLES
 	tristate "Netfilter Xtables support (required for ip_tables)"
 	default m if NETFILTER_ADVANCED=n
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index 238b6a620e88..06df49ea6329 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -128,6 +128,7 @@ obj-$(CONFIG_NFT_FWD_NETDEV)	+= nft_fwd_netdev.o
 obj-$(CONFIG_NF_FLOW_TABLE)	+= nf_flow_table.o
 nf_flow_table-objs		:= nf_flow_table_core.o nf_flow_table_ip.o \
 				   nf_flow_table_offload.o
+nf_flow_table-$(CONFIG_NF_FLOW_TABLE_PROCFS) += nf_flow_table_procfs.o
 
 obj-$(CONFIG_NF_FLOW_TABLE_INET) += nf_flow_table_inet.o
 
diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c
index f2def06d1070..51c2b1570838 100644
--- a/net/netfilter/nf_flow_table_core.c
+++ b/net/netfilter/nf_flow_table_core.c
@@ -614,14 +614,74 @@ void nf_flow_table_free(struct nf_flowtable *flow_table)
 }
 EXPORT_SYMBOL_GPL(nf_flow_table_free);
 
+static int nf_flow_table_init_net(struct net *net)
+{
+	net->ft.stat = alloc_percpu(struct nf_flow_table_stat);
+	return net->ft.stat ? 0 : -ENOMEM;
+}
+
+static void nf_flow_table_fini_net(struct net *net)
+{
+	free_percpu(net->ft.stat);
+}
+
+static int nf_flow_table_pernet_init(struct net *net)
+{
+	int ret;
+
+	ret = nf_flow_table_init_net(net);
+	if (ret < 0)
+		return ret;
+
+	ret = nf_flow_table_init_proc(net);
+	if (ret < 0)
+		goto out_proc;
+
+	return 0;
+
+out_proc:
+	nf_flow_table_fini_net(net);
+	return ret;
+}
+
+static void nf_flow_table_pernet_exit(struct list_head *net_exit_list)
+{
+	struct net *net;
+
+	list_for_each_entry(net, net_exit_list, exit_list) {
+		nf_flow_table_fini_proc(net);
+		nf_flow_table_fini_net(net);
+	}
+}
+
+static struct pernet_operations nf_flow_table_net_ops = {
+	.init = nf_flow_table_pernet_init,
+	.exit_batch = nf_flow_table_pernet_exit,
+};
+
 static int __init nf_flow_table_module_init(void)
 {
-	return nf_flow_table_offload_init();
+	int ret;
+
+	ret = register_pernet_subsys(&nf_flow_table_net_ops);
+	if (ret < 0)
+		return ret;
+
+	ret = nf_flow_table_offload_init();
+	if (ret)
+		goto out_offload;
+
+	return 0;
+
+out_offload:
+	unregister_pernet_subsys(&nf_flow_table_net_ops);
+	return ret;
 }
 
 static void __exit nf_flow_table_module_exit(void)
 {
 	nf_flow_table_offload_exit();
+	unregister_pernet_subsys(&nf_flow_table_net_ops);
 }
 
 module_init(nf_flow_table_module_init);
diff --git a/net/netfilter/nf_flow_table_offload.c b/net/netfilter/nf_flow_table_offload.c
index 11b6e1942092..103b6cbf257f 100644
--- a/net/netfilter/nf_flow_table_offload.c
+++ b/net/netfilter/nf_flow_table_offload.c
@@ -967,17 +967,22 @@ static void flow_offload_work_stats(struct flow_offload_work *offload)
 static void flow_offload_work_handler(struct work_struct *work)
 {
 	struct flow_offload_work *offload;
+	struct net *net;
 
 	offload = container_of(work, struct flow_offload_work, work);
+	net = read_pnet(&offload->flowtable->net);
 	switch (offload->cmd) {
 		case FLOW_CLS_REPLACE:
 			flow_offload_work_add(offload);
+			NF_FLOW_TABLE_STAT_DEC_ATOMIC(net, count_wq_add);
 			break;
 		case FLOW_CLS_DESTROY:
 			flow_offload_work_del(offload);
+			NF_FLOW_TABLE_STAT_DEC_ATOMIC(net, count_wq_del);
 			break;
 		case FLOW_CLS_STATS:
 			flow_offload_work_stats(offload);
+			NF_FLOW_TABLE_STAT_DEC_ATOMIC(net, count_wq_stats);
 			break;
 		default:
 			WARN_ON_ONCE(1);
@@ -989,12 +994,18 @@ static void flow_offload_work_handler(struct work_struct *work)
 
 static void flow_offload_queue_work(struct flow_offload_work *offload)
 {
-	if (offload->cmd == FLOW_CLS_REPLACE)
+	struct net *net = read_pnet(&offload->flowtable->net);
+
+	if (offload->cmd == FLOW_CLS_REPLACE) {
+		NF_FLOW_TABLE_STAT_INC(net, count_wq_add);
 		queue_work(nf_flow_offload_add_wq, &offload->work);
-	else if (offload->cmd == FLOW_CLS_DESTROY)
+	} else if (offload->cmd == FLOW_CLS_DESTROY) {
+		NF_FLOW_TABLE_STAT_INC(net, count_wq_del);
 		queue_work(nf_flow_offload_del_wq, &offload->work);
-	else
+	} else {
+		NF_FLOW_TABLE_STAT_INC(net, count_wq_stats);
 		queue_work(nf_flow_offload_stats_wq, &offload->work);
+	}
 }
 
 static struct flow_offload_work *
diff --git a/net/netfilter/nf_flow_table_procfs.c b/net/netfilter/nf_flow_table_procfs.c
new file mode 100644
index 000000000000..159b033a43e6
--- /dev/null
+++ b/net/netfilter/nf_flow_table_procfs.c
@@ -0,0 +1,80 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include <linux/kernel.h>
+#include <linux/proc_fs.h>
+#include <net/netfilter/nf_flow_table.h>
+
+static void *nf_flow_table_cpu_seq_start(struct seq_file *seq, loff_t *pos)
+{
+	struct net *net = seq_file_net(seq);
+	int cpu;
+
+	if (*pos == 0)
+		return SEQ_START_TOKEN;
+
+	for (cpu = *pos - 1; cpu < nr_cpu_ids; ++cpu) {
+		if (!cpu_possible(cpu))
+			continue;
+		*pos = cpu + 1;
+		return per_cpu_ptr(net->ft.stat, cpu);
+	}
+
+	return NULL;
+}
+
+static void *nf_flow_table_cpu_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+	struct net *net = seq_file_net(seq);
+	int cpu;
+
+	for (cpu = *pos; cpu < nr_cpu_ids; ++cpu) {
+		if (!cpu_possible(cpu))
+			continue;
+		*pos = cpu + 1;
+		return per_cpu_ptr(net->ft.stat, cpu);
+	}
+	(*pos)++;
+	return NULL;
+}
+
+static void nf_flow_table_cpu_seq_stop(struct seq_file *seq, void *v)
+{
+}
+
+static int nf_flow_table_cpu_seq_show(struct seq_file *seq, void *v)
+{
+	const struct nf_flow_table_stat *st = v;
+
+	if (v == SEQ_START_TOKEN) {
+		seq_puts(seq, "wq_add   wq_del   wq_stats\n");
+		return 0;
+	}
+
+	seq_printf(seq, "%8d %8d %8d\n",
+		   st->count_wq_add,
+		   st->count_wq_del,
+		   st->count_wq_stats
+		);
+	return 0;
+}
+
+static const struct seq_operations nf_flow_table_cpu_seq_ops = {
+	.start	= nf_flow_table_cpu_seq_start,
+	.next	= nf_flow_table_cpu_seq_next,
+	.stop	= nf_flow_table_cpu_seq_stop,
+	.show	= nf_flow_table_cpu_seq_show,
+};
+
+int nf_flow_table_init_proc(struct net *net)
+{
+	struct proc_dir_entry *pde;
+
+	pde = proc_create_net("nf_flowtable", 0444, net->proc_net_stat,
+			      &nf_flow_table_cpu_seq_ops,
+			      sizeof(struct seq_net_private));
+	return pde ? 0 : -ENOMEM;
+}
+
+void nf_flow_table_fini_proc(struct net *net)
+{
+	remove_proc_entry("nf_flowtable", net->proc_net_stat);
+}