[PATCH net 0/2] net: Fix bridge enslavement failure

[PATCH net 0/2] net: Fix bridge enslavement failure

[Ido Schimmel]

From: Ido Schimmel <idosch@nvidia.com>

Patch #1 fixes an issue in which an upper netdev cannot be enslaved to a
bridge when it has multiple netdevs with different parent identifiers
beneath it.

Patch #2 adds a test case using two netdevsim instances.

Ido Schimmel (2):
  net: Fix bridge enslavement failure
  selftests: rtnetlink: Test bridge enslavement with different parent
    IDs

 net/core/dev.c                           |  2 +-
 tools/testing/selftests/net/rtnetlink.sh | 47 ++++++++++++++++++++++++
 2 files changed, 48 insertions(+), 1 deletion(-)

-- 
2.26.2

[PATCH net 1/2] net: Fix bridge enslavement failure

[Ido Schimmel]

From: Ido Schimmel <idosch@nvidia.com>

When a netdev is enslaved to a bridge, its parent identifier is queried.
This is done so that packets that were already forwarded in hardware
will not be forwarded again by the bridge device between netdevs
belonging to the same hardware instance.

The operation fails when the netdev is an upper of netdevs with
different parent identifiers.

Instead of failing the enslavement, have dev_get_port_parent_id() return
'-EOPNOTSUPP' which will signal the bridge to skip the query operation.
Other callers of the function are not affected by this change.

Fixes: 7e1146e8c10c ("net: devlink: introduce devlink_compat_switch_id_get() helper")
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reported-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <nikolay@nvidia.com>
---
 net/core/dev.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index 4086d335978c..266073e300b5 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -8647,7 +8647,7 @@ int dev_get_port_parent_id(struct net_device *dev,
 		if (!first.id_len)
 			first = *ppid;
 		else if (memcmp(&first, ppid, sizeof(*ppid)))
-			return -ENODATA;
+			return -EOPNOTSUPP;
 	}
 
 	return err;
-- 
2.26.2

[PATCH net 2/2] selftests: rtnetlink: Test bridge enslavement with different parent IDs

[Ido Schimmel]

From: Ido Schimmel <idosch@nvidia.com>

Test that an upper device of netdevs with different parent IDs can be
enslaved to a bridge.

The test fails without previous commit.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <nikolay@nvidia.com>
---
 tools/testing/selftests/net/rtnetlink.sh | 47 ++++++++++++++++++++++++
 1 file changed, 47 insertions(+)

diff --git a/tools/testing/selftests/net/rtnetlink.sh b/tools/testing/selftests/net/rtnetlink.sh
index 7c38a909f8b8..8a2fe6d64bf2 100755
--- a/tools/testing/selftests/net/rtnetlink.sh
+++ b/tools/testing/selftests/net/rtnetlink.sh
@@ -1175,6 +1175,51 @@ kci_test_neigh_get()
 	echo "PASS: neigh get"
 }
 
+kci_test_bridge_parent_id()
+{
+	local ret=0
+	sysfsnet=/sys/bus/netdevsim/devices/netdevsim
+	probed=false
+
+	if [ ! -w /sys/bus/netdevsim/new_device ] ; then
+		modprobe -q netdevsim
+		check_err $?
+		if [ $ret -ne 0 ]; then
+			echo "SKIP: bridge_parent_id can't load netdevsim"
+			return $ksft_skip
+		fi
+		probed=true
+	fi
+
+	echo "10 1" > /sys/bus/netdevsim/new_device
+	while [ ! -d ${sysfsnet}10 ] ; do :; done
+	echo "20 1" > /sys/bus/netdevsim/new_device
+	while [ ! -d ${sysfsnet}20 ] ; do :; done
+	udevadm settle
+	dev10=`ls ${sysfsnet}10/net/`
+	dev20=`ls ${sysfsnet}20/net/`
+
+	ip link add name test-bond0 type bond mode 802.3ad
+	ip link set dev $dev10 master test-bond0
+	ip link set dev $dev20 master test-bond0
+	ip link add name test-br0 type bridge
+	ip link set dev test-bond0 master test-br0
+	check_err $?
+
+	# clean up any leftovers
+	ip link del dev test-br0
+	ip link del dev test-bond0
+	echo 20 > /sys/bus/netdevsim/del_device
+	echo 10 > /sys/bus/netdevsim/del_device
+	$probed && rmmod netdevsim
+
+	if [ $ret -ne 0 ]; then
+		echo "FAIL: bridge_parent_id"
+		return 1
+	fi
+	echo "PASS: bridge_parent_id"
+}
+
 kci_test_rtnl()
 {
 	local ret=0
@@ -1224,6 +1269,8 @@ kci_test_rtnl()
 	check_err $?
 	kci_test_neigh_get
 	check_err $?
+	kci_test_bridge_parent_id
+	check_err $?
 
 	kci_del_dummy
 	return $ret
-- 
2.26.2

Re: [PATCH net 0/2] net: Fix bridge enslavement failure

[Jakub Kicinski]

On Thu, 10 Sep 2020 14:01:25 +0300 Ido Schimmel wrote:
> From: Ido Schimmel <idosch@nvidia.com>
> 
> Patch #1 fixes an issue in which an upper netdev cannot be enslaved to a
> bridge when it has multiple netdevs with different parent identifiers
> beneath it.
> 
> Patch #2 adds a test case using two netdevsim instances.

Awesome, thanks Ido!

Reviewed-by: Jakub Kicinski <kuba@kernel.org>

Re: [PATCH net 0/2] net: Fix bridge enslavement failure

[David Miller]

From: Ido Schimmel <idosch@idosch.org>
Date: Thu, 10 Sep 2020 14:01:25 +0300

> From: Ido Schimmel <idosch@nvidia.com>
> 
> Patch #1 fixes an issue in which an upper netdev cannot be enslaved to a
> bridge when it has multiple netdevs with different parent identifiers
> beneath it.
> 
> Patch #2 adds a test case using two netdevsim instances.

Series applied and patch #1 queued up for -stable, thank you.