kernel/linux.git/include/net/dsa.h, branch v6.6.131

net: dsa: properly keep track of conduit reference

2026-03-25T10:06:00+00:00

commit 06e219f6a706c367c93051f408ac61417643d2f9 upstream. Problem description ------------------- DSA has a mumbo-jumbo of reference handling of the conduit net device and its kobject which, sadly, is just wrong and doesn't make sense. There are two distinct problems. 1. The OF path, which uses of_find_net_device_by_node(), never releases the elevated refcount on the conduit's kobject. Nominally, the OF and non-OF paths should result in objects having identical reference counts taken, and it is already suspicious that dsa_dev_to_net_device() has a put_device() call which is missing in dsa_port_parse_of(), but we can actually even verify that an issue exists. With CONFIG_DEBUG_KOBJECT_RELEASE=y, if we run this command "before" and "after" applying this patch: (unbind the conduit driver for net device eno2) echo 0000:00:00.2 > /sys/bus/pci/drivers/fsl_enetc/unbind we see these lines in the output diff which appear only with the patch applied: kobject: 'eno2' (ffff002009a3a6b8): kobject_release, parent 0000000000000000 (delayed 1000) kobject: '109' (ffff0020099d59a0): kobject_release, parent 0000000000000000 (delayed 1000) 2. After we find the conduit interface one way (OF) or another (non-OF), it can get unregistered at any time, and DSA remains with a long-lived, but in this case stale, cpu_dp->conduit pointer. Holding the net device's underlying kobject isn't actually of much help, it just prevents it from being freed (but we never need that kobject directly). What helps us to prevent the net device from being unregistered is the parallel netdev reference mechanism (dev_hold() and dev_put()). Actually we actually use that netdev tracker mechanism implicitly on user ports since commit 2f1e8ea726e9 ("net: dsa: link interfaces with the DSA master to get rid of lockdep warnings"), via netdev_upper_dev_link(). But time still passes at DSA switch probe time between the initial of_find_net_device_by_node() code and the user port creation time, time during which the conduit could unregister itself and DSA wouldn't know about it. So we have to run of_find_net_device_by_node() under rtnl_lock() to prevent that from happening, and release the lock only with the netdev tracker having acquired the reference. Do we need to keep the reference until dsa_unregister_switch() / dsa_switch_shutdown()? 1: Maybe yes. A switch device will still be registered even if all user ports failed to probe, see commit 86f8b1c01a0a ("net: dsa: Do not make user port errors fatal"), and the cpu_dp->conduit pointers remain valid. I haven't audited all call paths to see whether they will actually use the conduit in lack of any user port, but if they do, it seems safer to not rely on user ports for that reference. 2. Definitely yes. We support changing the conduit which a user port is associated to, and we can get into a situation where we've moved all user ports away from a conduit, thus no longer hold any reference to it via the net device tracker. But we shouldn't let it go nonetheless - see the next change in relation to dsa_tree_find_first_conduit() and LAG conduits which disappear. We have to be prepared to return to the physical conduit, so the CPU port must explicitly keep another reference to it. This is also to say: the user ports and their CPU ports may not always keep a reference to the same conduit net device, and both are needed. As for the conduit's kobject for the /sys/class/net/ entry, we don't care about it, we can release it as soon as we hold the net device object itself. History and blame attribution ----------------------------- The code has been refactored so many times, it is very difficult to follow and properly attribute a blame, but I'll try to make a short history which I hope to be correct. We have two distinct probing paths: - one for OF, introduced in 2016 in commit 83c0afaec7b7 ("net: dsa: Add new binding implementation") - one for non-OF, introduced in 2017 in commit 71e0bbde0d88 ("net: dsa: Add support for platform data") These are both complete rewrites of the original probing paths (which used struct dsa_switch_driver and other weird stuff, instead of regular devices on their respective buses for register access, like MDIO, SPI, I2C etc): - one for OF, introduced in 2013 in commit 5e95329b701c ("dsa: add device tree bindings to register DSA switches") - one for non-OF, introduced in 2008 in commit 91da11f870f0 ("net: Distributed Switch Architecture protocol support") except for tiny bits and pieces like dsa_dev_to_net_device() which were seemingly carried over since the original commit, and used to this day. The point is that the original probing paths received a fix in 2015 in the form of commit 679fb46c5785 ("net: dsa: Add missing master netdev dev_put() calls"), but the fix never made it into the "new" (dsa2) probing paths that can still be traced to today, and the fixed probing path was later deleted in 2019 in commit 93e86b3bc842 ("net: dsa: Remove legacy probing support"). That is to say, the new probing paths were never quite correct in this area. The existence of the legacy probing support which was deleted in 2019 explains why dsa_dev_to_net_device() returns a conduit with elevated refcount (because it was supposed to be released during dsa_remove_dst()). After the removal of the legacy code, the only user of dsa_dev_to_net_device() calls dev_put(conduit) immediately after this function returns. This pattern makes no sense today, and can only be interpreted historically to understand why dev_hold() was there in the first place. Change details -------------- Today we have a better netdev tracking infrastructure which we should use. Logically netdev_hold() belongs in common code (dsa_port_parse_cpu(), where dp->conduit is assigned), but there is a tradeoff to be made with the rtnl_lock() section which would become a bit too long if we did that - dsa_port_parse_cpu() also calls request_module(). So we duplicate a bit of logic in order for the callers of dsa_port_parse_cpu() to be the ones responsible of holding the conduit reference and releasing it on error. This shortens the rtnl_lock() section significantly. In the dsa_switch_probe() error path, dsa_switch_release_ports() will be called in a number of situations, one being where dsa_port_parse_cpu() maybe didn't get the chance to run at all (a different port failed earlier, etc). So we have to test for the conduit being NULL prior to calling netdev_put(). There have still been so many transformations to the code since the blamed commits (rename master -> conduit, commit 0650bf52b31f ("net: dsa: be compatible with masters which unregister on shutdown")), that it only makes sense to fix the code using the best methods available today and see how it can be backported to stable later. I suspect the fix cannot even be backported to kernels which lack dsa_switch_shutdown(), and I suspect this is also maybe why the long-lived conduit reference didn't make it into the new DSA probing paths at the time (problems during shutdown). Because dsa_dev_to_net_device() has a single call site and has to be changed anyway, the logic was just absorbed into the non-OF dsa_port_parse(). Tested on the ocelot/felix switch and on dsa_loop, both on the NXP LS1028A with CONFIG_DEBUG_KOBJECT_RELEASE=y. Reported-by: Ma Ke Closes: https://lore.kernel.org/netdev/20251214131204.4684-1-make24@iscas.ac.cn/ Fixes: 83c0afaec7b7 ("net: dsa: Add new binding implementation") Fixes: 71e0bbde0d88 ("net: dsa: Add support for platform data") Reviewed-by: Jonas Gorski Signed-off-by: Vladimir Oltean Link: https://patch.msgid.link/20251215150236.3931670-1-vladimir.oltean@nxp.com Signed-off-by: Paolo Abeni [ backport: "conduit" -> "master" in code, kept original commit message ] Signed-off-by: Vladimir Oltean Signed-off-by: Greg Kroah-Hartman

net: dsa: remove legacy_pre_march2020 detection

2023-07-18T07:47:08+00:00

All drivers are now updated for the March 2020 changes, and no longer make use of the mac_pcs_get_state() or mac_an_restart() operations, which are now NULL across all DSA drivers. All DSA drivers don't look at speed, duplex, pause or advertisement in their phylink_mac_config() method either. Remove support for these operations from DSA, and stop marking DSA as a legacy driver by default. Signed-off-by: Russell King (Oracle) Reviewed-by: Florian Fainelli Reviewed-by: Vladimir Oltean Signed-off-by: Paolo Abeni

Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

2023-06-27T16:45:22+00:00

Merge in late fixes to prepare for the 6.5 net-next PR. Signed-off-by: Jakub Kicinski

net: dsa: avoid suspicious RCU usage for synced VLAN-aware MAC addresses

2023-06-27T16:37:41+00:00

When using the felix driver (the only one which supports UC filtering and MC filtering) as a DSA master for a random other DSA switch, one can see the following stack trace when the downstream switch ports join a VLAN-aware bridge: ============================= WARNING: suspicious RCU usage ----------------------------- net/8021q/vlan_core.c:238 suspicious rcu_dereference_protected() usage! stack backtrace: Workqueue: dsa_ordered dsa_slave_switchdev_event_work Call trace: lockdep_rcu_suspicious+0x170/0x210 vlan_for_each+0x8c/0x188 dsa_slave_sync_uc+0x128/0x178 __hw_addr_sync_dev+0x138/0x158 dsa_slave_set_rx_mode+0x58/0x70 __dev_set_rx_mode+0x88/0xa8 dev_uc_add+0x74/0xa0 dsa_port_bridge_host_fdb_add+0xec/0x180 dsa_slave_switchdev_event_work+0x7c/0x1c8 process_one_work+0x290/0x568 What it's saying is that vlan_for_each() expects rtnl_lock() context and it's not getting it, when it's called from the DSA master's ndo_set_rx_mode(). The caller of that - dsa_slave_set_rx_mode() - is the slave DSA interface's dsa_port_bridge_host_fdb_add() which comes from the deferred dsa_slave_switchdev_event_work(). We went to great lengths to avoid the rtnl_lock() context in that call path in commit 0faf890fc519 ("net: dsa: drop rtnl_lock from dsa_slave_switchdev_event_work"), and calling rtnl_lock() is simply not an option due to the possibility of deadlocking when calling dsa_flush_workqueue() from the call paths that do hold rtnl_lock() - basically all of them. So, when the DSA master calls vlan_for_each() from its ndo_set_rx_mode(), the state of the 8021q driver on this device is really not protected from concurrent access by anything. Looking at net/8021q/, I don't think that vlan_info->vid_list was particularly designed with RCU traversal in mind, so introducing an RCU read-side form of vlan_for_each() - vlan_for_each_rcu() - won't be so easy, and it also wouldn't be exactly what we need anyway. In general I believe that the solution isn't in net/8021q/ anyway; vlan_for_each() is not cut out for this task. DSA doesn't need rtnl_lock() to be held per se - since it's not a netdev state change that we're blocking, but rather, just concurrent additions/removals to a VLAN list. We don't even need sleepable context - the callback of vlan_for_each() just schedules deferred work. The proposed escape is to remove the dependency on vlan_for_each() and to open-code a non-sleepable, rtnl-free alternative to that, based on copies of the VLAN list modified from .ndo_vlan_rx_add_vid() and .ndo_vlan_rx_kill_vid(). Fixes: 64fdc5f341db ("net: dsa: sync unicast and multicast addresses for VLAN filters too") Signed-off-by: Vladimir Oltean Link: https://lore.kernel.org/r/20230626154402.3154454-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski

Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

2023-06-23T01:40:38+00:00

Cross-merge networking fixes after downstream PR. Conflicts: tools/testing/selftests/net/fcnal-test.sh d7a2fc1437f7 ("selftests: net: fcnal-test: check if FIPS mode is enabled") dd017c72dde6 ("selftests: fcnal: Test SO_DONTROUTE on TCP sockets.") https://lore.kernel.org/all/5007b52c-dd16-dbf6-8d64-b9701bfa498b@tessares.net/ https://lore.kernel.org/all/20230619105427.4a0df9b3@canb.auug.org.au/ No adjacent changes. Signed-off-by: Jakub Kicinski

net: dsa: introduce preferred_default_local_cpu_port and use on MT7530

2023-06-20T08:40:26+00:00

Since the introduction of the OF bindings, DSA has always had a policy that in case multiple CPU ports are present in the device tree, the numerically smallest one is always chosen. The MT7530 switch family, except the switch on the MT7988 SoC, has 2 CPU ports, 5 and 6, where port 6 is preferable on the MT7531BE switch because it has higher bandwidth. The MT7530 driver developers had 3 options: - to modify DSA when the MT7531 switch support was introduced, such as to prefer the better port - to declare both CPU ports in device trees as CPU ports, and live with the sub-optimal performance resulting from not preferring the better port - to declare just port 6 in the device tree as a CPU port Of course they chose the path of least resistance (3rd option), kicking the can down the road. The hardware description in the device tree is supposed to be stable - developers are not supposed to adopt the strategy of piecemeal hardware description, where the device tree is updated in lockstep with the features that the kernel currently supports. Now, as a result of the fact that they did that, any attempts to modify the device tree and describe both CPU ports as CPU ports would make DSA change its default selection from port 6 to 5, effectively resulting in a performance degradation visible to users with the MT7531BE switch as can be seen below. Without preferring port 6: [ ID][Role] Interval Transfer Bitrate Retr [ 5][TX-C] 0.00-20.00 sec 374 MBytes 157 Mbits/sec 734 sender [ 5][TX-C] 0.00-20.00 sec 373 MBytes 156 Mbits/sec receiver [ 7][RX-C] 0.00-20.00 sec 1.81 GBytes 778 Mbits/sec 0 sender [ 7][RX-C] 0.00-20.00 sec 1.81 GBytes 777 Mbits/sec receiver With preferring port 6: [ ID][Role] Interval Transfer Bitrate Retr [ 5][TX-C] 0.00-20.00 sec 1.99 GBytes 856 Mbits/sec 273 sender [ 5][TX-C] 0.00-20.00 sec 1.99 GBytes 855 Mbits/sec receiver [ 7][RX-C] 0.00-20.00 sec 1.72 GBytes 737 Mbits/sec 15 sender [ 7][RX-C] 0.00-20.00 sec 1.71 GBytes 736 Mbits/sec receiver Using one port for WAN and the other ports for LAN is a very popular use case which is what this test emulates. As such, this change proposes that we retroactively modify stable kernels (which don't support the modification of the CPU port assignments, so as to let user space fix the problem and restore the throughput) to keep the mt7530 driver preferring port 6 even with device trees where the hardware is more fully described. Fixes: c288575f7810 ("net: dsa: mt7530: Add the support of MT7531 switch") Signed-off-by: Vladimir Oltean Signed-off-by: Arınç ÜNAL Reviewed-by: Russell King (Oracle) Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller

net: dsa: add support for mac_prepare() and mac_finish() calls

2023-05-26T09:39:40+00:00

Add DSA support for the phylink mac_prepare() and mac_finish() calls. These were introduced as part of the PCS support to allow MACs to perform preparatory steps prior to configuration, and finalisation steps after the MAC and PCS has been configured. Introducing phylink_pcs support to the mv88e6xxx DSA driver needs some code moved out of its mac_config() stage into the mac_prepare() and mac_finish() stages, and this commit facilitates such code in DSA drivers. Reviewed-by: Andrew Lunn Signed-off-by: Russell King (Oracle) Reviewed-by: Simon Horman Signed-off-by: David S. Miller

net: create a netdev notifier for DSA to reject PTP on DSA master

2023-04-03T09:04:27+00:00

The fact that PTP 2-step TX timestamping is broken on DSA switches if the master also timestamps the same packets is documented by commit f685e609a301 ("net: dsa: Deny PTP on master if switch supports it"). We attempt to help the users avoid shooting themselves in the foot by making DSA reject the timestamping ioctls on an interface that is a DSA master, and the switch tree beneath it contains switches which are aware of PTP. The only problem is that there isn't an established way of intercepting ndo_eth_ioctl calls, so DSA creates avoidable burden upon the network stack by creating a struct dsa_netdevice_ops with overlaid function pointers that are manually checked from the relevant call sites. There used to be 2 such dsa_netdevice_ops, but now, ndo_eth_ioctl is the only one left. There is an ongoing effort to migrate driver-visible hardware timestamping control from the ndo_eth_ioctl() based API to a new ndo_hwtstamp_set() model, but DSA actively prevents that migration, since dsa_master_ioctl() is currently coded to manually call the master's legacy ndo_eth_ioctl(), and so, whenever a network device driver would be converted to the new API, DSA's restrictions would be circumvented, because any device could be used as a DSA master. The established way for unrelated modules to react on a net device event is via netdevice notifiers. So we create a new notifier which gets called whenever there is an attempt to change hardware timestamping settings on a device. Finally, there is another reason why a netdev notifier will be a good idea, besides strictly DSA, and this has to do with PHY timestamping. With ndo_eth_ioctl(), all MAC drivers must manually call phy_has_hwtstamp() before deciding whether to act upon SIOCSHWTSTAMP, otherwise they must pass this ioctl to the PHY driver via phy_mii_ioctl(). With the new ndo_hwtstamp_set() API, it will be desirable to simply not make any calls into the MAC device driver when timestamping should be performed at the PHY level. But there exist drivers, such as the lan966x switch, which need to install packet traps for PTP regardless of whether they are the layer that provides the hardware timestamps, or the PHY is. That would be impossible to support with the new API. The proposal there, too, is to introduce a netdev notifier which acts as a better cue for switching drivers to add or remove PTP packet traps, than ndo_hwtstamp_set(). The one introduced here "almost" works there as well, except for the fact that packet traps should only be installed if the PHY driver succeeded to enable hardware timestamping, whereas here, we need to deny hardware timestamping on the DSA master before it actually gets enabled. This is why this notifier is called "PRE_", and the notifier that would get used for PHY timestamping and packet traps would be called NETDEV_CHANGE_HWTSTAMP. This isn't a new concept, for example NETDEV_CHANGEUPPER and NETDEV_PRECHANGEUPPER do the same thing. In expectation of future netlink UAPI, we also pass a non-NULL extack pointer to the netdev notifier, and we make DSA populate it with an informative reason for the rejection. To avoid making it go to waste, we make the ioctl-based dev_set_hwtstamp() create a fake extack and print the message to the kernel log. Link: https://lore.kernel.org/netdev/20230401191215.tvveoi3lkawgg6g4@skbuf/ Link: https://lore.kernel.org/netdev/20230310164451.ls7bbs6pdzs4m6pw@skbuf/ Signed-off-by: Vladimir Oltean Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller

net: dsa: add plumbing for changing and getting MAC merge layer state

2023-01-23T12:44:18+00:00

The DSA core is in charge of the ethtool_ops of the net devices associated with switch ports, so in case a hardware driver supports the MAC merge layer, DSA must pass the callbacks through to the driver. Add support for precisely that. Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller

net: dsa: move tag_8021q headers to their proper place

2022-11-23T04:41:53+00:00

tag_8021q definitions are all over the place. Some are exported to linux/dsa/8021q.h (visible by DSA core, taggers, switch drivers and everyone else), and some are in dsa_priv.h. Move the structures that don't need external visibility into tag_8021q.c, and the ones which don't need the world or switch drivers to see them into tag_8021q.h. We also have the tag_8021q.h inclusion from switch.c, which is basically the entire reason why tag_8021q.c was built into DSA in commit 8b6e638b4be2 ("net: dsa: build tag_8021q.c as part of DSA core"). I still don't know how to better deal with that, so leave it alone. Signed-off-by: Vladimir Oltean Reviewed-by: Florian Fainelli Signed-off-by: Jakub Kicinski