Extend lan966x with RGMII support. The MAC supports all RGMII_* modes.
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Link: https://lore.kernel.org/r/20220902111548.614525-1-horatiu.vultur@microchip.comSigned-off-by: NPaolo Abeni <pabeni@redhat.com>

There is no dedicated reset for just the switch core. The reset which
is used up until now, is more of a global reset, resetting almost the
whole SoC and cause spurious errors by doing so. Make it possible to
handle the reset elsewhere and make the reset optional.
Signed-off-by: NMichael Walle <michael@walle.cc>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Don't just print a warning.  Clean up and return an error as well.

Fixes: c8349639 ("net: lan966x: Add FDMA functionality")
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Link: https://lore.kernel.org/r/YwjgDm/SVd5c1tQU@kiliSigned-off-by: NJakub Kicinski <kuba@kernel.org>

Extend MAC support to support also lag interfaces:
1. In case an entry is learned on a port that is part of lag interface,
   then notify the upper layers that the entry is learned on the bond
   interface
2. If a port leaves the bond and the port is the first port in the lag
   group, then it is required to update all MAC entries to change the
   destination port.
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Offload FDB entries when the original device is a lag interface. Because
all the ports under the lag have the same chip id, which is the chip id
of first port, then add the entries only for the first port.
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add link aggregation hardware offload support for lan966x
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Extend lan966x_foreign_bridging_check to check also if the upper
interface is a lag device. Don't allow a lan966x port to be part of a
lag if it has foreign interfaces.
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Expose lan966x_switchdev_nb and lan966x_switchdev_blocking_nb to the
lan966x_main.h file because they will be needed by the lag driver.
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Whenever a port leaves a bridge, flush the workqueue of the FDB work.
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Split the function lan966x_fdb_event_work. One case for when the
orig_dev is a bridge and one case when orig_dev is lan966x port.
This is preparation for lag support. There is no functional change.
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add the registers used by lan966x to configure the lag interface.
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The Lan996x controller supports the QUSGMII mode, which is very similar
to QSGMII in the way it's configured and the autonegociation
capababilities it provides.

This commit adds support for that mode, treating it most of the time
like QSGMII, making sure that we do configure the PCS how we should.
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Reviewed-by: NAndrew Lunn <andrew@lunn.ch>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The platform_get_irq_byname() returns non-zero IRQ number
or negative error number. "if (irq)" always true, chang it
to "if (irq > 0)"
Signed-off-by: NLi Qiong <liqiong@nfschina.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When the SW bridge was trying to add/remove entries to/from HW, the
access to HW was not protected by any lock. In this way, it was
possible to have race conditions.
Fix this by using the lan966x->mac_lock to protect parallel access to HW
for this cases.

Fixes: 25ee9561 ("net: lan966x: More MAC table functionality")
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Reviewed-by: NVladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

The problem with this spin lock is that it was just protecting the list
of the MAC entries in SW and not also the access to the MAC entries in HW.
Because the access to HW is indirect, then it could happen to have race
conditions.
For example when SW introduced an entry in MAC table and the irq mac is
trying to read something from the MAC.
Update such that also the access to MAC entries in HW is protected by
this lock.

Fixes: 5ccd66e0 ("net: lan966x: add support for interrupts from analyzer")
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Reviewed-by: NVladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

To remove an entry to the MAC table, it is required first to setup the
entry and then issue a command for the MAC to forget the entry.
So if it happens for two threads to remove simultaneously an entry
in MAC table then it would be a race condition.
Fix this by using lan966x->mac_lock to protect the HW access.

Fixes: e18aba89 ("net: lan966x: add mactable support")
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Reviewed-by: NVladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

To add an entry to the MAC table, it is required first to setup the
entry and then issue a command for the MAC to learn the entry.
So if it happens for two threads to add simultaneously an entry in MAC
table then it would be a race condition.
Fix this by using lan966x->mac_lock to protect the HW access.

Fixes: fc0c3fe7 ("net: lan966x: Add function lan966x_mac_ip_learn()")
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Reviewed-by: NVladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

When the HW deletes an entry in MAC table then it generates an
interrupt. The SW will go through it's own list of MAC entries and if it
is not found then it would notify the listeners about this. The problem
is that when the SW will go through it's own list it would take a spin
lock(lan966x->mac_lock) and when it notifies that the entry is deleted.
But to notify the listeners it taking the rtnl_lock which is illegal.

This is fixed by instead of notifying right away that the entry is
deleted, move the entry on a temp list and once, it checks all the
entries then just notify that the entries from temp list are deleted.

Fixes: 5ccd66e0 ("net: lan966x: add support for interrupts from analyzer")
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Reviewed-by: NVladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

Instead of counting the child nodes in the device tree, hardcode the
number of ports in the driver itself.  The counting won't work at all
if an ethernet port is marked as disabled, e.g. because it is not
connected on the board at all.

It turns out that the LAN9662 and LAN9668 use the same switching IP
with the same synthesis parameters. The only difference is that the
output ports are not connected. Thus, we can just hardcode the
number of physical ports to 8.

Fixes: db8bcaad ("net: lan966x: add the basic lan966x driver")
Signed-off-by: NMichael Walle <michael@walle.cc>
Reviewed-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Link: https://lore.kernel.org/r/20220704153654.1167886-1-michael@walle.ccSigned-off-by: NJakub Kicinski <kuba@kernel.org>

At the moment, if devm_of_phy_get() returns an error the serdes
simply isn't set. While it is bad to ignore an error in general, there
is a particular bug that network isn't working if the serdes driver is
compiled as a module. In that case, devm_of_phy_get() returns
-EDEFER_PROBE and the error is silently ignored.

The serdes is optional, it is not there if the port is using RGMII, in
which case devm_of_phy_get() returns -ENODEV. Rearrange the error
handling so that -ENODEV will be handled but other error codes will
abort the probing.

Fixes: d28d6d2e ("net: lan966x: add port module support")
Signed-off-by: NMichael Walle <michael@walle.cc>
Link: https://lore.kernel.org/r/20220525231239.1307298-1-michael@walle.ccSigned-off-by: NJakub Kicinski <kuba@kernel.org>

The following two scenarios were failing for lan966x.
1. If the port had the address X and then trying to assign the same
   address, then the HW was just removing this address because first it
   tries to learn new address and then delete the old one. As they are
   the same the HW remove it.
2. If the port eth0 was assigned the same address as one of the other
   ports eth1 then when assigning back the address to eth0 then the HW
   was deleting the address of eth1.

The case 1. is fixed by checking if the port has already the same
address while case 2. is fixed by checking if the address is used by any
other port.

Fixes: e18aba89 ("net: lan966x: add mactable support")
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Link: https://lore.kernel.org/r/20220513180030.3076793-1-horatiu.vultur@microchip.comSigned-off-by: NPaolo Abeni <pabeni@redhat.com>

The smatch found the following warning:

drivers/net/ethernet/microchip/lan966x/lan966x_fdma.c:736 lan966x_fdma_reload()
warn: 'rx_dcbs' was already freed.

This issue can happen when changing the MTU on one of the ports and once
the RX buffers are allocated and then the TX buffer allocation fails.
In that case the RX buffers should not be restore. This fix this issue
such that the RX buffers will not be restored if the TX buffers failed
to be allocated.

Fixes: 2ea1cbac ("net: lan966x: Update FDMA to change MTU.")
Reported-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Link: https://lore.kernel.org/r/20220511204059.2689199-1-horatiu.vultur@microchip.comSigned-off-by: NJakub Kicinski <kuba@kernel.org>

The PHY subsystem as well as the MIIM mdio driver (in case of the
integrated PHYs) will take care of the resets. A separate reset driver
isn't needed. There is no in-tree user of this feature. Remove the
support.
Signed-off-by: NMichael Walle <michael@walle.cc>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Extend the PTP programmable pins to implement also PTP_PF_EXTTS
function. The PTP pin can be configured to capture only on the rising
edge of the PPS signal. And once an event is seen then an interrupt is
generated and the local time counter is saved.
The interrupt is shared between all the pins.
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Lan966x has 8 PTP programmable pins, where the last pins is hardcoded to
be used by PHC0, which does the frame timestamping. All the rest of the
PTP pins can be shared between the PHCs and can have different functions
like perout or extts. For now add support for PTP_FS_PEROUT.
The HW is not able to support absolute start time but can use the nsec
for phase adjustment when generating PPS.
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add registers that are used to configure the PTP pins. These registers
are used to enable the interrupts per PTP pin and to set the waveform
generated by the pin.
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

To read/write a value to a PHC, it is required to use a PTP pin.
Currently it is used pin 5, but change to pin 7 as is the last pin.
All the other pins will have different functions.
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The lan966x->ports[] array has lan966x->num_phys_ports elements.  These
are assigned in lan966x_probe().  That means the > comparison should be
changed to >=.

The first off by one check is harmless but the second one could lead to
an out of bounds access and a crash.

Fixes: 5ccd66e0 ("net: lan966x: add support for interrupts from analyzer")
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When the lan966x driver is removed make sure to remove also the ptp_irq
IRQ.

Fixes: e85a96e4 ("net: lan966x: Add support for ptp interrupts")
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Link: https://lore.kernel.org/r/20220413195716.3796467-1-horatiu.vultur@microchip.comSigned-off-by: NJakub Kicinski <kuba@kernel.org>

When changing the MTU, it is required to change also the size of the
DBs. In case those frames will arrive to CPU.
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

Ethernet frames can be extracted or injected to or from the device's
DDR memory. There is one channel for injection and one channel for
extraction. Each of these channels contain a linked list of DCBs which
contains DB. The DCB contains only 1 DB for both the injection and
extraction. Each DB contains a frame. Every time when a frame is received
or transmitted an interrupt is generated.

It is not possible to use both the FDMA and the manual
injection/extraction of the frames. Therefore the FDMA has priority over
the manual because of better performance values.

FDMA:
iperf -c 192.168.1.1
[  5]   0.00-10.02  sec   420 MBytes   352 Mbits/sec    0 sender
[  5]   0.00-10.03  sec   420 MBytes   351 Mbits/sec      receiver

iperf -c 192.168.1.1 -R
[  5]   0.00-10.01  sec   528 MBytes   442 Mbits/sec    0 sender
[  5]   0.00-10.00  sec   524 MBytes   440 Mbits/sec      receiver

Manual:
iperf -c 192.168.1.1
[  5]   0.00-10.02  sec  93.8 MBytes  78.5 Mbits/sec    0 sender
[  5]   0.00-10.03  sec  93.8 MBytes  78.4 Mbits/sec      receiver

ipers -c 192.168.1.1 -R
[  5]   0.00-10.03  sec   121 MBytes   101 Mbits/sec    0 sender
[  5]   0.00-10.01  sec   118 MBytes  99.0 Mbits/sec      receiver
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

Expose the following functions 'lan966x_hw_offload',
'lan966x_ifh_get_src_port' and 'lan966x_ifh_get_timestamp' in
lan966x_main.h so they can be accessed by FDMA.
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

Add the registers that are used to configure the FDMA.
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

Currently when getting a new MAC is learn, the HW generates an
interrupt. So then the SW will check the new entry and checks if it
arrived on a correct port. If it didn't just generate a warning.
But this could still crash the system. Therefore stop processing that
entry when an issue is seen.

Fixes: 5ccd66e0 ("net: lan966x: add support for interrupts from analyzer")
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

On lan966x it is not allowed to have foreign interfaces under a bridge
which already contains lan966x ports. So when a port leaves the bridge
it would call switchdev_bridge_port_unoffload which eventually will
notify the other ports that bridge left the vlan group but that is not
true because the bridge is still part of the vlan group.

Therefore when a port leaves the bridge, stop generating replays because
already the HW cleared after itself and the other ports don't need to do
anything else.

Fixes: cf2f6089 ("net: lan966x: Add support to offload the forwarding.")
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

In case an IGMP frame has a vlan tag, then the function
lan966x_hw_offload couldn't figure out that is a IGMP frame. Therefore
the SW thinks that the frame was already forward by the HW which is not
true.
Extend lan966x_hw_offload to pop the vlan tag if are any and then check
for IGMP frames.

Fixes: 47aeea0d ("net: lan966x: Implement the callback SWITCHDEV_ATTR_ID_BRIDGE_MC_DISABLED ")
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

The clk_per_cfg register represents the value added to the system clock
for each clock cycle. The issue is that the default value is wrong,
meaning that in case the DUT was a grandmaster then everone in the
network was too slow. In case there was a grandmaster, then there is no
issue because the DUT will configure clk_per_cfg register based on the
master frequency.

Fixes: d0964594 ("net: lan966x: Add support for ptp clocks")
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NJakub Kicinski <kuba@kernel.org>

ioctls handled by phy_mii_ioctl() will cause a kernel oops when the
interface is down. Fix it by making sure there is a PHY attached.

Fixes: 735fec99 ("net: lan966x: Implement SIOCSHWTSTAMP and SIOCGHWTSTAMP")
Signed-off-by: NMichael Walle <michael@walle.cc>
Reviewed-by: NAndrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20220328220350.3118969-1-michael@walle.ccSigned-off-by: NPaolo Abeni <pabeni@redhat.com>

When doing manual injection of the frame, it is required to check if the
TX FIFO is ready to accept the next word of the frame. For this we are
using 'readx_poll_timeout_atomic', the only problem is that before it
actually checks the status, is determining the time when to finish polling
the status. Which seems to be an expensive operation.
Therefore check the status of the TX FIFO before calling
'readx_poll_timeout_atomic'.
Doing this will improve the TX bitrate by ~70%. Because 99% the FIFO is
ready by that time. The measurements were done using iperf3.

Before:
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-10.03  sec  55.2 MBytes  46.2 Mbits/sec    0 sender
[  5]   0.00-10.04  sec  53.8 MBytes  45.0 Mbits/sec      receiver

After:
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-10.10  sec  95.0 MBytes  78.9 Mbits/sec    0 sender
[  5]   0.00-10.11  sec  95.0 MBytes  78.8 Mbits/sec      receiver
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Reviewed-by: NAndrew Lunn <andrew@lunn.ch>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The registers used to inject a frame to one of the ports is shared
between all the net devices. Therefore, there can be race conditions for
accessing the registers when two processes send frames at the same time
on different ports.

To fix this, add a spinlock around the function
'lan966x_port_ifh_xmit()'.
Signed-off-by: NHoratiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>