commit 8fd54a73b7cda11548154451bdb4bde6d8ff74c7 upstream.
Since teardown is supposed to undo the effects of the setup method, it
should be called in the error path for dsa_switch_setup, not just in
dsa_switch_teardown.
Fixes: 5e3f847a02 ("net: dsa: Add teardown callback for drivers")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20210204163351.2929670-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 07b90056cb15ff9877dca0d8f1b6583d1051f724 upstream.
Currently the following happens when a DSA master driver unbinds while
there are DSA switches attached to it:
$ echo 0000:00:00.5 > /sys/bus/pci/drivers/mscc_felix/unbind
------------[ cut here ]------------
WARNING: CPU: 0 PID: 392 at net/core/dev.c:9507
Call trace:
rollback_registered_many+0x5fc/0x688
unregister_netdevice_queue+0x98/0x120
dsa_slave_destroy+0x4c/0x88
dsa_port_teardown.part.16+0x78/0xb0
dsa_tree_teardown_switches+0x58/0xc0
dsa_unregister_switch+0x104/0x1b8
felix_pci_remove+0x24/0x48
pci_device_remove+0x48/0xf0
device_release_driver_internal+0x118/0x1e8
device_driver_detach+0x28/0x38
unbind_store+0xd0/0x100
Located at the above location is this WARN_ON:
/* Notifier chain MUST detach us all upper devices. */
WARN_ON(netdev_has_any_upper_dev(dev));
Other stacked interfaces, like VLAN, do indeed listen for
NETDEV_UNREGISTER on the real_dev and also unregister themselves at that
time, which is clearly the behavior that rollback_registered_many
expects. But DSA interfaces are not VLAN. They have backing hardware
(platform devices, PCI devices, MDIO, SPI etc) which have a life cycle
of their own and we can't just trigger an unregister from the DSA
framework when we receive a netdev notifier that the master unregisters.
Luckily, there is something we can do, and that is to inform the driver
core that we have a runtime dependency to the DSA master interface's
device, and create a device link where that is the supplier and we are
the consumer. Having this device link will make the DSA switch unbind
before the DSA master unbinds, which is enough to avoid the WARN_ON from
rollback_registered_many.
Note that even before the blamed commit, DSA did nothing intelligent
when the master interface got unregistered either. See the discussion
here:
https://lore.kernel.org/netdev/20200505210253.20311-1-f.fainelli@gmail.com/
But this time, at least the WARN_ON is loud enough that the
upper_dev_link commit can be blamed.
The advantage with this approach vs dev_hold(master) in the attached
link is that the latter is not meant for long term reference counting.
With dev_hold, the only thing that will happen is that when the user
attempts an unbind of the DSA master, netdev_wait_allrefs will keep
waiting and waiting, due to DSA keeping the refcount forever. DSA would
not access freed memory corresponding to the master interface, but the
unbind would still result in a freeze. Whereas with device links,
graceful teardown is ensured. It even works with cascaded DSA trees.
$ echo 0000:00:00.2 > /sys/bus/pci/drivers/fsl_enetc/unbind
[ 1818.797546] device swp0 left promiscuous mode
[ 1819.301112] sja1105 spi2.0: Link is Down
[ 1819.307981] DSA: tree 1 torn down
[ 1819.312408] device eno2 left promiscuous mode
[ 1819.656803] mscc_felix 0000:00:00.5: Link is Down
[ 1819.667194] DSA: tree 0 torn down
[ 1819.711557] fsl_enetc 0000:00:00.2 eno2: Link is Down
This approach allows us to keep the DSA framework absolutely unchanged,
and the driver core will just know to unbind us first when the master
goes away - as opposed to the large (and probably impossible) rework
required if attempting to listen for NETDEV_UNREGISTER.
As per the documentation at Documentation/driver-api/device_link.rst,
specifying the DL_FLAG_AUTOREMOVE_CONSUMER flag causes the device link
to be automatically purged when the consumer fails to probe or later
unbinds. So we don't need to keep the consumer_link variable in struct
dsa_switch.
Fixes: 2f1e8ea726 ("net: dsa: link interfaces with the DSA master to get rid of lockdep warnings")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Tested-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20210111230943.3701806-1-olteanv@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 91158e1680b164c8d101144ca916a3dca10c3e17 ]
Florian reported a use-after-free bug in devlink_nl_port_fill found with
KASAN:
(devlink_nl_port_fill)
(devlink_port_notify)
(devlink_port_unregister)
(dsa_switch_teardown.part.3)
(dsa_tree_teardown_switches)
(dsa_unregister_switch)
(bcm_sf2_sw_remove)
(platform_remove)
(device_release_driver_internal)
(device_links_unbind_consumers)
(device_release_driver_internal)
(device_driver_detach)
(unbind_store)
Allocated by task 31:
alloc_netdev_mqs+0x5c/0x50c
dsa_slave_create+0x110/0x9c8
dsa_register_switch+0xdb0/0x13a4
b53_switch_register+0x47c/0x6dc
bcm_sf2_sw_probe+0xaa4/0xc98
platform_probe+0x90/0xf4
really_probe+0x184/0x728
driver_probe_device+0xa4/0x278
__device_attach_driver+0xe8/0x148
bus_for_each_drv+0x108/0x158
Freed by task 249:
free_netdev+0x170/0x194
dsa_slave_destroy+0xac/0xb0
dsa_port_teardown.part.2+0xa0/0xb4
dsa_tree_teardown_switches+0x50/0xc4
dsa_unregister_switch+0x124/0x250
bcm_sf2_sw_remove+0x98/0x13c
platform_remove+0x44/0x5c
device_release_driver_internal+0x150/0x254
device_links_unbind_consumers+0xf8/0x12c
device_release_driver_internal+0x84/0x254
device_driver_detach+0x30/0x34
unbind_store+0x90/0x134
What happens is that devlink_port_unregister emits a netlink
DEVLINK_CMD_PORT_DEL message which associates the devlink port that is
getting unregistered with the ifindex of its corresponding net_device.
Only trouble is, the net_device has already been unregistered.
It looks like we can stub out the search for a corresponding net_device
if we clear the devlink_port's type. This looks like a bit of a hack,
but also seems to be the reason why the devlink_port_type_clear function
exists in the first place.
Fixes: 3122433eb5 ("net: dsa: Register devlink ports before calling DSA driver setup()")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Tested-by: Florian fainelli <f.fainelli@gmail.com>
Reported-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20210112004831.3778323-1-olteanv@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Marvell 88E6060 uses tag_trailer.c and the KSZ8795, KSZ9477 and
KSZ9893 switches also use tail tags.
Fixes: 7a6ffe764b ("net: dsa: point out the tail taggers")
Signed-off-by: Christian Eggers <ceggers@arri.de>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20201016171603.10587-1-ceggers@arri.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Simplify the code by using new function dev_fetch_sw_netstats().
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Tested-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/b6047017-8226-6b7e-a3cd-064e69fdfa27@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When the Extraction Frame Header contains a valid classified VLAN, use
that instead of the VLAN header present in the packet.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
A driver may refuse to enable VLAN filtering for any reason beyond what
the DSA framework cares about, such as:
- having tc-flower rules that rely on the switch being VLAN-aware
- the particular switch does not support VLAN, even if the driver does
(the DSA framework just checks for the presence of the .port_vlan_add
and .port_vlan_del pointers)
- simply not supporting this configuration to be toggled at runtime
Currently, when a driver rejects a configuration it cannot support, it
does this from the commit phase, which triggers various warnings in
switchdev.
So propagate the prepare phase to drivers, to give them the ability to
refuse invalid configurations cleanly and avoid the warnings.
Since we need to modify all function prototypes and check for the
prepare phase from within the drivers, take that opportunity and move
the existing driver restrictions within the prepare phase where that is
possible and easy.
Cc: Florian Fainelli <f.fainelli@gmail.com>
Cc: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Cc: Hauke Mehrtens <hauke@hauke-m.de>
Cc: Woojung Huh <woojung.huh@microchip.com>
Cc: Microchip Linux Driver Support <UNGLinuxDriver@microchip.com>
Cc: Sean Wang <sean.wang@mediatek.com>
Cc: Landen Chao <Landen.Chao@mediatek.com>
Cc: Andrew Lunn <andrew@lunn.ch>
Cc: Vivien Didelot <vivien.didelot@gmail.com>
Cc: Jonathan McDowell <noodles@earth.li>
Cc: Linus Walleij <linus.walleij@linaro.org>
Cc: Alexandre Belloni <alexandre.belloni@bootlin.com>
Cc: Claudiu Manoil <claudiu.manoil@nxp.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Allow DSA drivers to make use of devlink port regions, via simple
wrappers.
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Tested-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
DSA drivers want to create regions on devlink ports as well as the
devlink device instance, in order to export registers and other tables
per port. To keep all this code together in the drivers, have the
devlink ports registered early, so the setup() method can setup both
device and port devlink regions.
v3:
Remove dp->setup
Move common code out of switch statement.
Fix wrong goto
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Tested-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If a port is unused, still create a devlink port for it, but set the
flavour to unused. This allows us to attach devlink regions to the
port, etc.
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Tested-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now that we are guaranteed that dsa_untag_bridge_pvid() is called after
eth_type_trans() we can utilize __vlan_find_dev_deep_rcu() which will
take care of finding an 802.1Q upper on top of a bridge master.
A common use case, prior to 12a1526d067 ("net: dsa: untag the bridge
pvid from rx skbs") was to configure a bridge 802.1Q upper like this:
ip link add name br0 type bridge vlan_filtering 0
ip link add link br0 name br0.1 type vlan id 1
in order to pop the default_pvid VLAN tag.
With this change we restore that behavior while still allowing the DSA
receive path to automatically pop the VLAN tag.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now that dsa_untag_bridge_pvid() is called after eth_type_trans() we are
guaranteed that skb->protocol will be set to a correct value, thus
allowing us to avoid calling vlan_eth_hdr().
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Indicate to the DSA receive path that we need to untage the bridge PVID,
this allows us to remove the dsa_untag_bridge_pvid() calls from
net/dsa/tag_brcm.c.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a DSA switch driver needs to call dsa_untag_bridge_pvid(), it can
set dsa_switch::untag_brige_pvid to indicate this is necessary.
This is a pre-requisite to making sure that we are always calling
dsa_untag_bridge_pvid() after eth_type_trans() has been called.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove the .flow_dissect procedure, so the flow dissector will call the
generic variant which works for this tagging protocol.
Cc: Linus Walleij <linus.walleij@linaro.org>
Cc: DENG Qingfang <dqfext@gmail.com>
Cc: Mauri Sandberg <sandberg@mailfence.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The sja1105 is a bit of a special snowflake, in that not all frames are
transmitted/received in the same way. L2 link-local frames are received
with the source port/switch ID information put in the destination MAC
address. For the rest, a tag_8021q header is used. So only the latter
frames displace the rest of the headers and need to use the generic flow
dissector procedure.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove the .flow_dissect procedure, so the flow dissector will call the
generic variant which works for this tagging protocol.
Cc: John Crispin <john@phrozen.org>
Cc: Alexander Lobakin <alobakin@pm.me>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove the .flow_dissect procedure, so the flow dissector will call the
generic variant which works for this tagging protocol.
Cc: DENG Qingfang <dqfext@gmail.com>
Cc: Sean Wang <sean.wang@mediatek.com>
Cc: John Crispin <john@phrozen.org>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove the .flow_dissect procedure, so the flow dissector will call the
generic variant which works for this tagging protocol.
Cc: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove the .flow_dissect procedure, so the flow dissector will call the
generic variant which works for this tagging protocol.
Cc: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are 2 Broadcom tags in use, one places the DSA tag before the
Ethernet destination MAC address, and the other before the EtherType.
Nonetheless, both displace the rest of the headers, so this tagger can
use the generic flow dissector procedure which accounts for that.
The ASCII art drawing is a good reference though, so keep it but move it
somewhere else.
Cc: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The Marvell 88E6060 uses tag_trailer.c and the KSZ8795, KSZ9477 and
KSZ9893 switches also use tail tags.
Tell that to the DSA core, since this makes a difference for the flow
dissector. Most switches break the parsing of frame headers, but these
ones don't, so no flow dissector adjustment needs to be done for them.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is no tagger that returns anything other than zero, so just change
the return type appropriately.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are 2 goals that we follow:
- Reduce the header size
- Make the header size equal between RX and TX
The issue that required long prefix on RX was the fact that the ocelot
DSA tag, being put before Ethernet as it is, would overlap with the area
that a DSA master uses for RX filtering (destination MAC address
mainly).
Now that we can ask DSA to put the master in promiscuous mode, in theory
we could remove the prefix altogether and call it a day, but it looks
like we can't. Using no prefix on ingress, some packets (such as ICMP)
would be received, while others (such as PTP) would not be received.
This is because the DSA master we use (enetc) triggers parse errors
("MAC rx frame errors") presumably because it sees Ethernet frames with
a bad length. And indeed, when using no prefix, the EtherType (bytes
12-13 of the frame, bits 96-111) falls over the REW_VAL field from the
extraction header, aka the PTP timestamp.
When turning the short (32-bit) prefix on, the EtherType overlaps with
bits 64-79 of the extraction header, which are a reserved area
transmitted as zero by the switch. The packets are not dropped by the
DSA master with a short prefix. Actually, the frames look like this in
tcpdump (below is a PTP frame, with an extra dsa_8021q tag - dadb 0482 -
added by a downstream sja1105).
89:0c:a9:f2:01:00 > 88:80:00:0a:00:1d, 802.3, length 0: LLC, \
dsap Unknown (0x10) Individual, ssap ProWay NM (0x0e) Response, \
ctrl 0x0004: Information, send seq 2, rcv seq 0, \
Flags [Response], length 78
0x0000: 8880 000a 001d 890c a9f2 0100 0000 100f ................
0x0010: 0400 0000 0180 c200 000e 001f 7b63 0248 ............{c.H
0x0020: dadb 0482 88f7 1202 0036 0000 0000 0000 .........6......
0x0030: 0000 0000 0000 0000 0000 001f 7bff fe63 ............{..c
0x0040: 0248 0001 1f81 0500 0000 0000 0000 0000 .H..............
0x0050: 0000 0000 0000 0000 0000 0000 ............
So the short prefix is our new default: we've shortened our RX frames by
12 octets, increased TX by 4, and headers are now equal between RX and
TX. Note that we still need promiscuous mode for the DSA master to not
drop it.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently PTP is broken when ports are in standalone mode (the tagger
keeps printing this message):
sja1105 spi0.1: Expected meta frame, is 01-80-c2-00-00-0e in the DSA master multicast filter?
Sure, one might say "simply add 01-80-c2-00-00-0e to the master's RX
filter" but things become more complicated because:
- Actually all frames in the 01-80-c2-xx-xx-xx and 01-1b-19-xx-xx-xx
range are trapped to the CPU automatically
- The switch mangles bytes 3 and 4 of the MAC address via the incl_srcpt
("include source port [in the DMAC]") option, which is how source port
and switch id identification is done for link-local traffic on RX. But
this means that an address installed to the RX filter would, at the
end of the day, not correspond to the final address seen by the DSA
master.
Assume RX filtering lists on DSA masters are typically too small to
include all necessary addresses for PTP to work properly on sja1105, and
just request promiscuous mode unconditionally.
Just an example:
Assuming the following addresses are trapped to the CPU:
01-80-c2-00-00-00 to 01-80-c2-00-00-ff
01-1b-19-00-00-00 to 01-1b-19-00-00-ff
These are 512 addresses.
Now let's say this is a board with 3 switches, and 4 ports per switch.
The 512 addresses become 6144 addresses that must be managed by the DSA
master's RX filtering lists.
This may be refined in the future, but for now, it is simply not worth
it to add the additional addresses to the master's RX filter, so simply
request it to become promiscuous as soon as the driver probes.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently DSA assumes that taggers don't mess with the destination MAC
address of the frames on RX. That is not always the case. Some DSA
headers are placed before the Ethernet header (ocelot), and others
simply mangle random bytes from the destination MAC address (sja1105
with its incl_srcpt option).
Currently the DSA master goes to promiscuous mode automatically when the
slave devices go too (such as when enslaved to a bridge), but in
standalone mode this is a problem that needs to be dealt with.
So give drivers the possibility to signal that their tagging protocol
will get randomly dropped otherwise, and let DSA deal with fixing that.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, ocelot switchdev passes the skb directly to the function that
enqueues it to the list of skb's awaiting a TX timestamp. Whereas the
felix DSA driver first clones the skb, then passes the clone to this
queue.
This matters because in the case of felix, the common IRQ handler, which
is ocelot_get_txtstamp(), currently clones the clone, and frees the
original clone. This is useless and can be simplified by using
skb_complete_tx_timestamp() instead of skb_tstamp_tx().
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Update the B53 driver to support VLANs while not filtering. This
requires us to enable VLAN globally within the switch upon driver
initial configuration (dev->vlan_enabled).
We also need to remove the code that dealt with PVID re-configuration in
b53_vlan_filtering() since that function worked under the assumption
that it would only be called to make a bridge VLAN filtering, or not
filtering, and we would attempt to move the port's PVID accordingly.
Now that VLANs are programmed all the time, even in the case of a
non-VLAN filtering bridge, we would be programming a default_pvid for
the bridged switch ports.
We need the DSA receive path to pop the VLAN tag if it is the bridge's
default_pvid because the CPU port is always programmed tagged in the
programmed VLANs. In order to do so we utilize the
dsa_untag_bridge_pvid() helper introduced in the commit before within
net/dsa/tag_brcm.c.
Acked-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently the bridge untags VLANs present in its VLAN groups in
__allowed_ingress() only when VLAN filtering is enabled.
But when a skb is seen on the RX path as tagged with the bridge's pvid,
and that bridge has vlan_filtering=0, and there isn't any 8021q upper
with that VLAN either, then we have a problem. The bridge will not untag
it (since it is supposed to remain VLAN-unaware), and pvid-tagged
communication will be broken.
There are 2 situations where we can end up like that:
1. When installing a pvid in egress-tagged mode, like this:
ip link add dev br0 type bridge vlan_filtering 0
ip link set swp0 master br0
bridge vlan del dev swp0 vid 1
bridge vlan add dev swp0 vid 1 pvid
This happens because DSA configures the VLAN membership of the CPU port
using the same flags as swp0 (in this case "pvid and not untagged"), in
an attempt to copy the frame as-is from ingress to the CPU.
However, in this case, the packet may arrive untagged on ingress, it
will be pvid-tagged by the ingress port, and will be sent as
egress-tagged towards the CPU. Otherwise stated, the CPU will see a VLAN
tag where there was none to speak of on ingress.
When vlan_filtering is 1, this is not a problem, as stated in the first
paragraph, because __allowed_ingress() will pop it. But currently, when
vlan_filtering is 0 and we have such a VLAN configuration, we need an
8021q upper (br0.1) to be able to ping over that VLAN, which is not
symmetrical with the vlan_filtering=1 case, and therefore, confusing for
users.
Basically what DSA attempts to do is simply an approximation: try to
copy the skb with (or without) the same VLAN all the way up to the CPU.
But DSA drivers treat CPU port VLAN membership in various ways (which is
a good segue into situation 2). And some of those drivers simply tell
the CPU port to copy the frame unmodified, which is the golden standard
when it comes to VLAN processing (therefore, any driver which can
configure the hardware to do that, should do that, and discard the VLAN
flags requested by DSA on the CPU port).
2. Some DSA drivers always configure the CPU port as egress-tagged, in
an attempt to recover the classified VLAN from the skb. These drivers
cannot work at all with untagged traffic when bridged in
vlan_filtering=0 mode. And they can't go for the easy "just keep the
pvid as egress-untagged towards the CPU" route, because each front port
can have its own pvid, and that might require conflicting VLAN
membership settings on the CPU port (swp1 is pvid for VID 1 and
egress-tagged for VID 2; swp2 is egress-taggeed for VID 1 and pvid for
VID 2; with this simplistic approach, the CPU port, which is really a
separate hardware entity and has its own VLAN membership settings, would
end up being egress-untagged in both VID 1 and VID 2, therefore losing
the VLAN tags of ingress traffic).
So the only thing we can do is to create a helper function for resolving
the problematic case (that is, a function which untags the bridge pvid
when that is in vlan_filtering=0 mode), which taggers in need should
call. It isn't called from the generic DSA receive path because there
are drivers that fall neither in the first nor second category.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Two minor conflicts:
1) net/ipv4/route.c, adding a new local variable while
moving another local variable and removing it's
initial assignment.
2) drivers/net/dsa/microchip/ksz9477.c, overlapping changes.
One pretty prints the port mode differently, whilst another
changes the driver to try and obtain the port mode from
the port node rather than the switch node.
Signed-off-by: David S. Miller <davem@davemloft.net>
Check whether there is any hwaccel VLAN tag on RX, and if there is,
treat it as the tag_8021q header.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The whole purpose of tag_8021q is to send VLAN-tagged traffic to the
CPU, from which the driver can decode the source port and switch id.
Currently this only works if the VLAN filtering on the master is
disabled. Change that by explicitly adding code to tag_8021q.c to add
the VLANs corresponding to the tags to the filter of the master
interface.
Because we now need to call vlan_vid_add, then we also need to hold the
RTNL mutex. Propagate that requirement to the callers of dsa_8021q_setup
and modify the existing call sites as appropriate. Note that one call
path, sja1105_best_effort_vlan_filtering_set -> sja1105_vlan_filtering
-> sja1105_setup_8021q_tagging, was already holding this lock.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Most DSA switch tags shift the EtherType to the right, causing the
master to not parse the VLAN as VLAN.
However, not all switches do that (example: tail tags, tag_8021q etc),
and if the DSA master has "rx-vlan-filter: on" in ethtool -k, then we
have a problem.
Therefore, we could populate the VLAN table of the master, just in case
(for some switches it will not make a difference), so that network I/O
can work even with a VLAN filtering master.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When the bridge has VLAN awareness disabled there isn't any duplication
of functionality, since the bridge does not process VLAN. Don't deny
adding 8021q uppers to DSA switch ports in that case. The switch is
supposed to simply pass traffic leaving the VLAN tag as-is, and the
stack will happily strip the VLAN tag for all 8021q uppers that exist.
We need to ensure that there are no 8021q uppers when the user attempts
to enable bridge vlan_filtering.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The current logic beats me a little bit. The comment that "bridge skips
-EOPNOTSUPP, so skip the prepare phase" was introduced in commit
fb2dabad69 ("net: dsa: support VLAN filtering switchdev attr").
I'm not sure:
(a) ok, the bridge skips -EOPNOTSUPP, but, so what, where are we
returning -EOPNOTSUPP?
(b) even if we are, and I'm just not seeing it, what is the causality
relationship between the bridge skipping -EOPNOTSUPP and DSA
skipping the prepare phase, and just returning zero?
One thing is certain beyond doubt though, and that is that DSA currently
refuses VLAN filtering from the "commit" phase instead of "prepare", and
that this is not a good thing:
ip link add br0 type bridge
ip link add br1 type bridge vlan_filtering 1
ip link set swp2 master br0
ip link set swp3 master br1
[ 3790.379389] 001: sja1105 spi0.1: VLAN filtering is a global setting
[ 3790.379399] 001: ------------[ cut here ]------------
[ 3790.379403] 001: WARNING: CPU: 1 PID: 515 at net/switchdev/switchdev.c:157 switchdev_port_attr_set_now+0x9c/0xa4
[ 3790.379420] 001: swp3: Commit of attribute (id=6) failed.
[ 3790.379533] 001: [<c11ff588>] (switchdev_port_attr_set_now) from [<c11b62e4>] (nbp_vlan_init+0x84/0x148)
[ 3790.379544] 001: [<c11b62e4>] (nbp_vlan_init) from [<c11a2ff0>] (br_add_if+0x514/0x670)
[ 3790.379554] 001: [<c11a2ff0>] (br_add_if) from [<c1031b5c>] (do_setlink+0x38c/0xab0)
[ 3790.379565] 001: [<c1031b5c>] (do_setlink) from [<c1036fe8>] (__rtnl_newlink+0x44c/0x748)
[ 3790.379573] 001: [<c1036fe8>] (__rtnl_newlink) from [<c1037328>] (rtnl_newlink+0x44/0x60)
[ 3790.379580] 001: [<c1037328>] (rtnl_newlink) from [<c10315fc>] (rtnetlink_rcv_msg+0x124/0x2f8)
[ 3790.379590] 001: [<c10315fc>] (rtnetlink_rcv_msg) from [<c10926b8>] (netlink_rcv_skb+0xb8/0x110)
[ 3790.379806] 001: ---[ end trace 0000000000000002 ]---
[ 3790.379819] 001: sja1105 spi0.1 swp3: failed to initialize vlan filtering on this port
So move the current logic that may fail (except ds->ops->port_vlan_filtering,
that is way harder) into the prepare stage of the switchdev transaction.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is checking for the following order of operations, and makes sure
to deny that configuration:
ip link add link swp2 name swp2.100 type vlan id 100
ip link add br0 type bridge vlan_filtering 1
ip link set swp2 master br0
bridge vlan add dev swp2 vid 100
Instead of using vlan_for_each(), which looks at the VLAN filters
installed with vlan_vid_add(), just track the 8021q uppers. This has the
advantage of freeing up the vlan_vid_add() call for actual VLAN
filtering.
There is another change in this patch. The check is moved in slave.c,
from switch.c. I don't think it makes sense to have this 8021q upper
check for each switch port that gets notified of that VLAN addition
(these include DSA links and CPU ports, we know those can't have 8021q
uppers because they don't have a net_device registered for them), so
just do it in slave.c, for that one slave interface.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
DSA tries to prevent having a VLAN added by a bridge and by an 802.1Q
upper at the same time. It does that by checking the VID in
.ndo_vlan_rx_add_vid(), since that's something that the 8021q module
calls, via vlan_vid_add(). When a VLAN matches in both subsystems, this
check returns -EBUSY.
However the vlan_vid_add() function isn't specific to the 8021q module
in any way at all. It is simply the kernel's way to tell an interface to
add a VLAN to its RX filter and not drop that VLAN. So there's no reason
to return -EBUSY when somebody tries to call vlan_vid_add() for a VLAN
that was installed by the bridge. The proper behavior is to accept that
configuration.
So what's wrong is how DSA checks that it has an 8021q upper. It should
look at the actual uppers for that, not just assume that the 8021q
module was somewhere in the call stack of .ndo_vlan_rx_add_vid().
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We'll be adding a new check in the PRECHANGEUPPER notifier, where we'll
need to check some VLAN uppers. It is hard to do that when there is
already a function named dsa_slave_upper_vlan_check. So rename this one.
Not to mention that this function probably shouldn't have started with
"dsa_slave_" in the first place, since the struct net_device argument
isn't a DSA slave, but an 8021q upper of one.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There doesn't seem to be any strong technical reason for doing it this
way, but we'll be adding more checks for invalid upper device
configurations, and it will be easier to have them all grouped under
PRECHANGEUPPER.
Tested that it still works:
ip link set br0 type bridge vlan_filtering 1
ip link add link swp2 name swp2.100 type vlan id 100
ip link set swp2.100 master br0
[ 20.321312] br0: port 5(swp2.100) entered blocking state
[ 20.326711] br0: port 5(swp2.100) entered disabled state
Error: dsa_core: Cannot enslave VLAN device into VLAN aware bridge.
[ 20.346549] br0: port 5(swp2.100) entered blocking state
[ 20.351957] br0: port 5(swp2.100) entered disabled state
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Allow the DSA drivers to implement the devlink call to get info info,
e.g. driver name, firmware version, ASIC ID, etc.
v2:
Combine declaration and the assignment on a single line.
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Allow DSA drivers to make use of devlink regions, via simple wrappers.
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Given a devlink instance, return the dsa switch it is associated to.
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ocelot_port->ts_id is used to:
(a) populate skb->cb[0] for matching the TX timestamp in the PTP IRQ
with an skb.
(b) populate the REW_OP from the injection header of the ongoing skb.
Only then is ocelot_port->ts_id incremented.
This is a problem because, at least theoretically, another timestampable
skb might use the same ocelot_port->ts_id before that is incremented.
Normally all transmit calls are serialized by the netdev transmit
spinlock, but in this case, ocelot_port_add_txtstamp_skb() is also
called by DSA, which has started declaring the NETIF_F_LLTX feature
since commit 2b86cb8299 ("net: dsa: declare lockless TX feature for
slave ports"). So the logic of using and incrementing the timestamp id
should be atomic per port.
The solution is to use the global ocelot_port->ts_id only while
protected by the associated ocelot_port->ts_id_lock. That's where we
populate skb->cb[0]. Note that for ocelot, ocelot_port_add_txtstamp_skb
is called for the actual skb, but for felix, it is called for the skb's
clone. That is something which will also be changed in the future.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Tested-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Reviewed-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This reverts commit 314f76d7a6.
Citing that commit message, the call graph was:
dsa_slave_vlan_rx_add_vid dsa_port_setup_8021q_tagging
| |
| |
| +-------------+
| |
v v
dsa_port_vid_add dsa_slave_port_obj_add
| |
+-------+ +-------+
| |
v v
dsa_port_vlan_add
Now that tag_8021q has its own ops structure, it no longer relies on
dsa_port_vid_add, and therefore on the dsa_switch_ops to install its
VLANs.
So dsa_port_vid_add now only has one single caller. So we can simplify
the call graph to what it was before, aka:
dsa_slave_vlan_rx_add_vid dsa_slave_port_obj_add
| |
+-------+ +-------+
| |
v v
dsa_port_vlan_add
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
While working on another tag_8021q driver implementation, some things
became apparent:
- It is not mandatory for a DSA driver to offload the tag_8021q VLANs by
using the VLAN table per se. For example, it can add custom TCAM rules
that simply encapsulate RX traffic, and redirect & decapsulate rules
for TX traffic. For such a driver, it makes no sense to receive the
tag_8021q configuration through the same callback as it receives the
VLAN configuration from the bridge and the 8021q modules.
- Currently, sja1105 (the only tag_8021q user) sets a
priv->expect_dsa_8021q variable to distinguish between the bridge
calling, and tag_8021q calling. That can be improved, to say the
least.
- The crosschip bridging operations are, in fact, stateful already. The
list of crosschip_links must be kept by the caller and passed to the
relevant tag_8021q functions.
So it would be nice if the tag_8021q configuration was more
self-contained. This patch attempts to do that.
Create a struct dsa_8021q_context which encapsulates a struct
dsa_switch, and has 2 function pointers for adding and deleting a VLAN.
These will replace the previous channel to the driver, which was through
the .port_vlan_add and .port_vlan_del callbacks of dsa_switch_ops.
Also put the list of crosschip_links into this dsa_8021q_context.
Drivers that don't support cross-chip bridging can simply omit to
initialize this list, as long as they dont call any cross-chip function.
The sja1105_vlan_add and sja1105_vlan_del functions are refactored into
a smaller sja1105_vlan_add_one, which now has 2 entry points:
- sja1105_vlan_add, from struct dsa_switch_ops
- sja1105_dsa_8021q_vlan_add, from the tag_8021q ops
But even this change is fairly trivial. It just reflects the fact that
for sja1105, the VLANs from these 2 channels end up in the same hardware
table. However that is not necessarily true in the general sense (and
that's the reason for making this change).
The rest of the patch is mostly plain refactoring of "ds" -> "ctx". The
dsa_8021q_context structure needs to be propagated because adding a VLAN
is now done through the ops function pointers inside of it.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is no point in calling dsa_port_setup_8021q_tagging for each
individual port. Additionally, it will become more difficult to do that
when we'll have a context structure to tag_8021q (next patch). So
refactor this now.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since commit 845e0ebb44 ("net: change addr_list_lock back to static
key"), cascaded DSA setups (DSA switch port as DSA master for another
DSA switch port) are emitting this lockdep warning:
============================================
WARNING: possible recursive locking detected
5.8.0-rc1-00133-g923e4b5032dd-dirty #208 Not tainted
--------------------------------------------
dhcpcd/323 is trying to acquire lock:
ffff000066dd4268 (&dsa_master_addr_list_lock_key/1){+...}-{2:2}, at: dev_mc_sync+0x44/0x90
but task is already holding lock:
ffff00006608c268 (&dsa_master_addr_list_lock_key/1){+...}-{2:2}, at: dev_mc_sync+0x44/0x90
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(&dsa_master_addr_list_lock_key/1);
lock(&dsa_master_addr_list_lock_key/1);
*** DEADLOCK ***
May be due to missing lock nesting notation
3 locks held by dhcpcd/323:
#0: ffffdbd1381dda18 (rtnl_mutex){+.+.}-{3:3}, at: rtnl_lock+0x24/0x30
#1: ffff00006614b268 (_xmit_ETHER){+...}-{2:2}, at: dev_set_rx_mode+0x28/0x48
#2: ffff00006608c268 (&dsa_master_addr_list_lock_key/1){+...}-{2:2}, at: dev_mc_sync+0x44/0x90
stack backtrace:
Call trace:
dump_backtrace+0x0/0x1e0
show_stack+0x20/0x30
dump_stack+0xec/0x158
__lock_acquire+0xca0/0x2398
lock_acquire+0xe8/0x440
_raw_spin_lock_nested+0x64/0x90
dev_mc_sync+0x44/0x90
dsa_slave_set_rx_mode+0x34/0x50
__dev_set_rx_mode+0x60/0xa0
dev_mc_sync+0x84/0x90
dsa_slave_set_rx_mode+0x34/0x50
__dev_set_rx_mode+0x60/0xa0
dev_set_rx_mode+0x30/0x48
__dev_open+0x10c/0x180
__dev_change_flags+0x170/0x1c8
dev_change_flags+0x2c/0x70
devinet_ioctl+0x774/0x878
inet_ioctl+0x348/0x3b0
sock_do_ioctl+0x50/0x310
sock_ioctl+0x1f8/0x580
ksys_ioctl+0xb0/0xf0
__arm64_sys_ioctl+0x28/0x38
el0_svc_common.constprop.0+0x7c/0x180
do_el0_svc+0x2c/0x98
el0_sync_handler+0x9c/0x1b8
el0_sync+0x158/0x180
Since DSA never made use of the netdev API for describing links between
upper devices and lower devices, the dev->lower_level value of a DSA
switch interface would be 1, which would warn when it is a DSA master.
We can use netdev_upper_dev_link() to describe the relationship between
a DSA slave and a DSA master. To be precise, a DSA "slave" (switch port)
is an "upper" to a DSA "master" (host port). The relationship is "many
uppers to one lower", like in the case of VLAN. So, for that reason, we
use the same function as VLAN uses.
There might be a chance that somebody will try to take hold of this
interface and use it immediately after register_netdev() and before
netdev_upper_dev_link(). To avoid that, we do the registration and
linkage while holding the RTNL, and we use the RTNL-locked cousin of
register_netdev(), which is register_netdevice().
Since this warning was not there when lockdep was using dynamic keys for
addr_list_lock, we are blaming the lockdep patch itself. The network
stack _has_ been using static lockdep keys before, and it _is_ likely
that stacked DSA setups have been triggering these lockdep warnings
since forever, however I can't test very old kernels on this particular
stacked DSA setup, to ensure I'm not in fact introducing regressions.
Fixes: 845e0ebb44 ("net: change addr_list_lock back to static key")
Suggested-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 72579e14a1 ("net: dsa: don't fail to probe if we couldn't set
the MTU") changed, for some reason, the "err && err != -EOPNOTSUPP"
check into a simple "err". This causes the MTU warning to be printed
even for drivers that don't have the MTU operations implemented.
Fix that.
Signed-off-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
slave_dev->name is only populated at this stage if it was specified
through a label in the device tree. However that is not mandatory.
When it isn't, the error message looks like this:
[ 5.037057] fsl_enetc 0000:00:00.2 eth2: error -19 setting up slave PHY for eth%d
[ 5.044672] fsl_enetc 0000:00:00.2 eth2: error -19 setting up slave PHY for eth%d
[ 5.052275] fsl_enetc 0000:00:00.2 eth2: error -19 setting up slave PHY for eth%d
[ 5.059877] fsl_enetc 0000:00:00.2 eth2: error -19 setting up slave PHY for eth%d
which is especially confusing since the error gets printed on behalf of
the DSA master (fsl_enetc in this case).
Printing an error message that contains a valid reference to the DSA
port's name is difficult at this point in the initialization stage, so
at least we should print some info that is more reliable, even if less
user-friendly. That may be the driver name and the hardware port index.
After this change, the error is printed as:
[ 6.051587] mscc_felix 0000:00:00.5: error -19 setting up PHY for tree 0, switch 0, port 0
[ 6.061192] mscc_felix 0000:00:00.5: error -19 setting up PHY for tree 0, switch 0, port 1
[ 6.070765] mscc_felix 0000:00:00.5: error -19 setting up PHY for tree 0, switch 0, port 2
[ 6.080324] mscc_felix 0000:00:00.5: error -19 setting up PHY for tree 0, switch 0, port 3
Signed-off-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
