Merge branch 'drop_monitor-Convert-to-use-devlink-tracepoint'

Ido Schimmel says:

====================
drop_monitor: Convert to use devlink tracepoint

Drop monitor is able to monitor both software and hardware originated
drops. Software drops are monitored by having drop monitor register its
probe on the 'kfree_skb' tracepoint. Hardware originated drops are
monitored by having devlink call into drop monitor whenever it receives
a dropped packet from the underlying hardware.

This patch set converts drop monitor to monitor both software and
hardware originated drops in the same way - by registering its probe on
the relevant tracepoint.

In addition to drop monitor being more consistent, it is now also
possible to build drop monitor as module instead of as a builtin and
still monitor hardware originated drops. Initially, CONFIG_NET_DEVLINK
implied CONFIG_NET_DROP_MONITOR, but after commit def2fbff
("kconfig: allow symbols implied by y to become m") we can have
CONFIG_NET_DEVLINK=y and CONFIG_NET_DROP_MONITOR=m and hardware
originated drops will not be monitored.

Patch set overview:

Patch #1 adds a tracepoint in devlink for trap reports.

Patch #2 prepares probe functions in drop monitor for the new
tracepoint.

Patch #3 converts drop monitor to use the new tracepoint.

Patches #4-#6 perform cleanups after the conversion.

Patch #7 adds a test case for drop monitor. Both software originated
drops and hardware originated drops (using netdevsim) are tested.

Tested:

| CONFIG_NET_DEVLINK | CONFIG_NET_DROP_MONITOR | Build | SW drops | HW drops |
| -------------------|-------------------------|-------|----------|----------|
|          y         |            y            |   v   |     v    |     v    |
|          y         |            m            |   v   |     v    |     v    |
|          y         |            n            |   v   |     x    |     x    |
|          n         |            y            |   v   |     v    |     x    |
|          n         |            m            |   v   |     v    |     x    |
|          n         |            n            |   v   |     x    |     x    |
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

MAINTAINERS

@@ -12065,7 +12065,6 @@ M:	Neil Horman <nhorman@tuxdriver.com>
 L:	netdev@vger.kernel.org
 S:	Maintained
 W:	https://fedorahosted.org/dropwatch/
-F:	include/net/drop_monitor.h
 F:	include/uapi/linux/net_dropmon.h
 F:	net/core/drop_monitor.c
 

include/net/devlink.h

@@ -624,6 +624,22 @@ struct devlink_health_reporter_ops {
 		    struct netlink_ext_ack *extack);
 };
 
+/**
+ * struct devlink_trap_metadata - Packet trap metadata.
+ * @trap_name: Trap name.
+ * @trap_group_name: Trap group name.
+ * @input_dev: Input netdevice.
+ * @fa_cookie: Flow action user cookie.
+ * @trap_type: Trap type.
+ */
+struct devlink_trap_metadata {
+	const char *trap_name;
+	const char *trap_group_name;
+	struct net_device *input_dev;
+	const struct flow_action_cookie *fa_cookie;
+	enum devlink_trap_type trap_type;
+};
+
 /**
  * struct devlink_trap_policer - Immutable packet trap policer attributes.
  * @id: Policer identifier.

include/net/drop_monitor.h

-/* SPDX-License-Identifier: GPL-2.0-only */
-
-#ifndef _NET_DROP_MONITOR_H_
-#define _NET_DROP_MONITOR_H_
-
-#include <linux/ktime.h>
-#include <linux/netdevice.h>
-#include <linux/skbuff.h>
-#include <net/flow_offload.h>
-
-/**
- * struct net_dm_hw_metadata - Hardware-supplied packet metadata.
- * @trap_group_name: Hardware trap group name.
- * @trap_name: Hardware trap name.
- * @input_dev: Input netdevice.
- * @fa_cookie: Flow action user cookie.
- */
-struct net_dm_hw_metadata {
-	const char *trap_group_name;
-	const char *trap_name;
-	struct net_device *input_dev;
-	const struct flow_action_cookie *fa_cookie;
-};
-
-#if IS_REACHABLE(CONFIG_NET_DROP_MONITOR)
-void net_dm_hw_report(struct sk_buff *skb,
-		      const struct net_dm_hw_metadata *hw_metadata);
-#else
-static inline void
-net_dm_hw_report(struct sk_buff *skb,
-		 const struct net_dm_hw_metadata *hw_metadata)
-{
-}
-#endif
-
-#endif /* _NET_DROP_MONITOR_H_ */

include/trace/events/devlink.h

@@ -171,6 +171,43 @@ TRACE_EVENT(devlink_health_reporter_state_update,
 		  __entry->new_state)
 );
 
+/*
+ * Tracepoint for devlink packet trap:
+ */
+TRACE_EVENT(devlink_trap_report,
+	TP_PROTO(const struct devlink *devlink, struct sk_buff *skb,
+		 const struct devlink_trap_metadata *metadata),
+
+	TP_ARGS(devlink, skb, metadata),
+
+	TP_STRUCT__entry(
+		__string(bus_name, devlink->dev->bus->name)
+		__string(dev_name, dev_name(devlink->dev))
+		__string(driver_name, devlink->dev->driver->name)
+		__string(trap_name, metadata->trap_name)
+		__string(trap_group_name, metadata->trap_group_name)
+		__dynamic_array(char, input_dev_name, IFNAMSIZ)
+	),
+
+	TP_fast_assign(
+		struct net_device *input_dev = metadata->input_dev;
+
+		__assign_str(bus_name, devlink->dev->bus->name);
+		__assign_str(dev_name, dev_name(devlink->dev));
+		__assign_str(driver_name, devlink->dev->driver->name);
+		__assign_str(trap_name, metadata->trap_name);
+		__assign_str(trap_group_name, metadata->trap_group_name);
+		__assign_str(input_dev_name,
+			     (input_dev ? input_dev->name : "NULL"));
+	),
+
+	TP_printk("bus_name=%s dev_name=%s driver_name=%s trap_name=%s "
+		  "trap_group_name=%s input_dev_name=%s", __get_str(bus_name),
+		  __get_str(dev_name), __get_str(driver_name),
+		  __get_str(trap_name), __get_str(trap_group_name),
+		  __get_str(input_dev_name))
+);
+
 #endif /* _TRACE_DEVLINK_H */
 
 /* This part must be outside protection */

net/Kconfig

@@ -434,7 +434,6 @@ config NET_SOCK_MSG
 config NET_DEVLINK
 	bool
 	default n
-	imply NET_DROP_MONITOR
 
 config PAGE_POOL
 	bool

net/core/devlink.c

@@ -27,7 +27,6 @@
 #include <net/net_namespace.h>
 #include <net/sock.h>
 #include <net/devlink.h>
-#include <net/drop_monitor.h>
 #define CREATE_TRACE_POINTS
 #include <trace/events/devlink.h>
 
@@ -84,6 +83,7 @@ EXPORT_SYMBOL(devlink_dpipe_header_ipv6);
 
 EXPORT_TRACEPOINT_SYMBOL_GPL(devlink_hwmsg);
 EXPORT_TRACEPOINT_SYMBOL_GPL(devlink_hwerr);
+EXPORT_TRACEPOINT_SYMBOL_GPL(devlink_trap_report);
 
 static const struct nla_policy devlink_function_nl_policy[DEVLINK_PORT_FUNCTION_ATTR_MAX + 1] = {
 	[DEVLINK_PORT_FUNCTION_ATTR_HW_ADDR] = { .type = NLA_BINARY },
@@ -9261,20 +9261,19 @@ devlink_trap_stats_update(struct devlink_stats __percpu *trap_stats,
 }
 
 static void
-devlink_trap_report_metadata_fill(struct net_dm_hw_metadata *hw_metadata,
+devlink_trap_report_metadata_set(struct devlink_trap_metadata *metadata,
 				 const struct devlink_trap_item *trap_item,
 				 struct devlink_port *in_devlink_port,
 				 const struct flow_action_cookie *fa_cookie)
 {
-	struct devlink_trap_group_item *group_item = trap_item->group_item;
-
-	hw_metadata->trap_group_name = group_item->group->name;
-	hw_metadata->trap_name = trap_item->trap->name;
-	hw_metadata->fa_cookie = fa_cookie;
+	metadata->trap_name = trap_item->trap->name;
+	metadata->trap_group_name = trap_item->group_item->group->name;
+	metadata->fa_cookie = fa_cookie;
+	metadata->trap_type = trap_item->trap->type;
 
 	spin_lock(&in_devlink_port->type_lock);
 	if (in_devlink_port->type == DEVLINK_PORT_TYPE_ETH)
-		hw_metadata->input_dev = in_devlink_port->type_dev;
+		metadata->input_dev = in_devlink_port->type_dev;
 	spin_unlock(&in_devlink_port->type_lock);
 }
 
@@ -9292,21 +9291,17 @@ void devlink_trap_report(struct devlink *devlink, struct sk_buff *skb,
 
 {
 	struct devlink_trap_item *trap_item = trap_ctx;
-	struct net_dm_hw_metadata hw_metadata = {};
 
 	devlink_trap_stats_update(trap_item->stats, skb->len);
 	devlink_trap_stats_update(trap_item->group_item->stats, skb->len);
 
-	/* Control packets were not dropped by the device or encountered an
-	 * exception during forwarding and therefore should not be reported to
-	 * the kernel's drop monitor.
-	 */
-	if (trap_item->trap->type == DEVLINK_TRAP_TYPE_CONTROL)
-		return;
+	if (trace_devlink_trap_report_enabled()) {
+		struct devlink_trap_metadata metadata = {};
 
-	devlink_trap_report_metadata_fill(&hw_metadata, trap_item,
+		devlink_trap_report_metadata_set(&metadata, trap_item,
 						 in_devlink_port, fa_cookie);
-	net_dm_hw_report(skb, &hw_metadata);
+		trace_devlink_trap_report(devlink, skb, &metadata);
+	}
 }
 EXPORT_SYMBOL_GPL(devlink_trap_report);
 

tools/testing/selftests/net/Makefile

@@ -19,6 +19,7 @@ TEST_PROGS += txtimestamp.sh
 TEST_PROGS += vrf-xfrm-tests.sh
 TEST_PROGS += rxtimestamp.sh
 TEST_PROGS += devlink_port_split.py
+TEST_PROGS += drop_monitor_tests.sh
 TEST_PROGS_EXTENDED := in_netns.sh
 TEST_GEN_FILES =  socket nettest
 TEST_GEN_FILES += psock_fanout psock_tpacket msg_zerocopy reuseport_addr_any

tools/testing/selftests/net/config

@@ -30,3 +30,6 @@ CONFIG_NET_SCH_ETF=m
 CONFIG_NET_SCH_NETEM=y
 CONFIG_TEST_BLACKHOLE_DEV=m
 CONFIG_KALLSYMS=y
+CONFIG_TRACEPOINTS=y
+CONFIG_NET_DROP_MONITOR=m
+CONFIG_NETDEVSIM=m

tools/testing/selftests/net/drop_monitor_tests.sh

+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+# This test is for checking drop monitor functionality.
+
+ret=0
+# Kselftest framework requirement - SKIP code is 4.
+ksft_skip=4
+
+# all tests in this script. Can be overridden with -t option
+TESTS="
+	sw_drops
+	hw_drops
+"
+
+IP="ip -netns ns1"
+TC="tc -netns ns1"
+DEVLINK="devlink -N ns1"
+NS_EXEC="ip netns exec ns1"
+NETDEVSIM_PATH=/sys/bus/netdevsim/
+DEV_ADDR=1337
+DEV=netdevsim${DEV_ADDR}
+DEVLINK_DEV=netdevsim/${DEV}
+
+log_test()
+{
+	local rc=$1
+	local expected=$2
+	local msg="$3"
+
+	if [ ${rc} -eq ${expected} ]; then
+		printf "    TEST: %-60s  [ OK ]\n" "${msg}"
+		nsuccess=$((nsuccess+1))
+	else
+		ret=1
+		nfail=$((nfail+1))
+		printf "    TEST: %-60s  [FAIL]\n" "${msg}"
+	fi
+}
+
+setup()
+{
+	modprobe netdevsim &> /dev/null
+
+	set -e
+	ip netns add ns1
+	$IP link add dummy10 up type dummy
+
+	$NS_EXEC echo "$DEV_ADDR 1" > ${NETDEVSIM_PATH}/new_device
+	udevadm settle
+	local netdev=$($NS_EXEC ls ${NETDEVSIM_PATH}/devices/${DEV}/net/)
+	$IP link set dev $netdev up
+
+	set +e
+}
+
+cleanup()
+{
+	$NS_EXEC echo "$DEV_ADDR" > ${NETDEVSIM_PATH}/del_device
+	ip netns del ns1
+}
+
+sw_drops_test()
+{
+	echo
+	echo "Software drops test"
+
+	setup
+
+	local dir=$(mktemp -d)
+
+	$TC qdisc add dev dummy10 clsact
+	$TC filter add dev dummy10 egress pref 1 handle 101 proto ip \
+		flower dst_ip 192.0.2.10 action drop
+
+	$NS_EXEC mausezahn dummy10 -a 00:11:22:33:44:55 -b 00:aa:bb:cc:dd:ee \
+		-A 192.0.2.1 -B 192.0.2.10 -t udp sp=12345,dp=54321 -c 0 -q \
+		-d 100msec &
+	timeout 5 dwdump -o sw -w ${dir}/packets.pcap
+	(( $(tshark -r ${dir}/packets.pcap \
+		-Y 'ip.dst == 192.0.2.10' 2> /dev/null | wc -l) != 0))
+	log_test $? 0 "Capturing active software drops"
+
+	rm ${dir}/packets.pcap
+
+	{ kill %% && wait %%; } 2>/dev/null
+	timeout 5 dwdump -o sw -w ${dir}/packets.pcap
+	(( $(tshark -r ${dir}/packets.pcap \
+		-Y 'ip.dst == 192.0.2.10' 2> /dev/null | wc -l) == 0))
+	log_test $? 0 "Capturing inactive software drops"
+
+	rm -r $dir
+
+	cleanup
+}
+
+hw_drops_test()
+{
+	echo
+	echo "Hardware drops test"
+
+	setup
+
+	local dir=$(mktemp -d)
+
+	$DEVLINK trap set $DEVLINK_DEV trap blackhole_route action trap
+	timeout 5 dwdump -o hw -w ${dir}/packets.pcap
+	(( $(tshark -r ${dir}/packets.pcap \
+		-Y 'net_dm.hw_trap_name== blackhole_route' 2> /dev/null \
+		| wc -l) != 0))
+	log_test $? 0 "Capturing active hardware drops"
+
+	rm ${dir}/packets.pcap
+
+	$DEVLINK trap set $DEVLINK_DEV trap blackhole_route action drop
+	timeout 5 dwdump -o hw -w ${dir}/packets.pcap
+	(( $(tshark -r ${dir}/packets.pcap \
+		-Y 'net_dm.hw_trap_name== blackhole_route' 2> /dev/null \
+		| wc -l) == 0))
+	log_test $? 0 "Capturing inactive hardware drops"
+
+	rm -r $dir
+
+	cleanup
+}
+
+################################################################################
+# usage
+
+usage()
+{
+	cat <<EOF
+usage: ${0##*/} OPTS
+
+        -t <test>   Test(s) to run (default: all)
+                    (options: $TESTS)
+EOF
+}
+
+################################################################################
+# main
+
+while getopts ":t:h" opt; do
+	case $opt in
+		t) TESTS=$OPTARG;;
+		h) usage; exit 0;;
+		*) usage; exit 1;;
+	esac
+done
+
+if [ "$(id -u)" -ne 0 ];then
+	echo "SKIP: Need root privileges"
+	exit $ksft_skip;
+fi
+
+if [ ! -x "$(command -v ip)" ]; then
+	echo "SKIP: Could not run test without ip tool"
+	exit $ksft_skip
+fi
+
+if [ ! -x "$(command -v devlink)" ]; then
+	echo "SKIP: Could not run test without devlink tool"
+	exit $ksft_skip
+fi
+
+if [ ! -x "$(command -v tshark)" ]; then
+	echo "SKIP: Could not run test without tshark tool"
+	exit $ksft_skip
+fi
+
+if [ ! -x "$(command -v dwdump)" ]; then
+	echo "SKIP: Could not run test without dwdump tool"
+	exit $ksft_skip
+fi
+
+if [ ! -x "$(command -v udevadm)" ]; then
+	echo "SKIP: Could not run test without udevadm tool"
+	exit $ksft_skip
+fi
+
+if [ ! -x "$(command -v timeout)" ]; then
+	echo "SKIP: Could not run test without timeout tool"
+	exit $ksft_skip
+fi
+
+if [ ! -x "$(command -v mausezahn)" ]; then
+	echo "SKIP: Could not run test without mausezahn tool"
+	exit $ksft_skip
+fi
+
+tshark -G fields 2> /dev/null | grep -q net_dm
+if [ $? -ne 0 ]; then
+	echo "SKIP: tshark too old, missing net_dm dissector"
+	exit $ksft_skip
+fi
+
+# start clean
+cleanup &> /dev/null
+
+for t in $TESTS
+do
+	case $t in
+	sw_drops|sw)			sw_drops_test;;
+	hw_drops|hw)			hw_drops_test;;
+
+	help) echo "Test names: $TESTS"; exit 0;;
+	esac
+done
+
+if [ "$TESTS" != "none" ]; then
+	printf "\nTests passed: %3d\n" ${nsuccess}
+	printf "Tests failed: %3d\n"   ${nfail}
+fi
+
+exit $ret