This should be done in the device driver instead of the device tree.
Signed-off-by: Christian Eggers <ceggers@arri.de>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

This should be done in the device driver instead of the device tree.
Signed-off-by: Christian Eggers <ceggers@arri.de>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

The dsa.yaml device tree binding allows "ethernet-ports" (preferred) and
"ports".
Signed-off-by: Christian Eggers <ceggers@arri.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Convert the bindings document for Microchip KSZ Series Ethernet switches
from txt to yaml. Removed spi-cpha and spi-cpol flags is this should be
handled by the device driver.
Signed-off-by: Christian Eggers <ceggers@arri.de>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Yunsheng Lin says:

====================
Add an assert in napi_consume_skb()

This patch introduces a lockdep_assert_in_softirq() interface and
uses it to assert the case when napi_consume_skb() is not called in
the softirq context.
====================

Link: https://lore.kernel.org/r/1606214969-97849-1-git-send-email-linyunsheng@huawei.comSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Use napi_consume_skb() to assert the case when it is not called
in a atomic softirq context.
Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

The current semantic for napi_consume_skb() is that caller need
to provide non-zero budget when calling from NAPI context, and
breaking this semantic will cause hard to debug problem, because
_kfree_skb_defer() need to run in atomic context in order to push
the skb to the particular cpu' napi_alloc_cache atomically.

So add the lockdep_assert_in_softirq() to assert when the running
context is not in_softirq, in_softirq means softirq is serving or
BH is disabled, which has a ambiguous semantics due to the BH
disabled confusion, so add a comment to emphasize that.

And the softirq context can be interrupted by hard IRQ or NMI
context, lockdep_assert_in_softirq() need to assert about hard
IRQ or NMI context too.
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

These are only used as input arguments to qmi_handle_init() which
accepts const pointers to both qmi_ops and qmi_msg_handler. Make them
const to allow the compiler to put them in read-only memory.
Signed-off-by: Rikard Falkeborn <rikard.falkeborn@gmail.com>
Acked-by: Alex Elder <elder@linaro.org>
Link: https://lore.kernel.org/r/20201122234031.33432-2-rikard.falkeborn@gmail.comSigned-off-by: Jakub Kicinski <kuba@kernel.org>

We can enter the main mptcp_recvmsg() loop even when
no subflows are connected. As note by Eric, that would
result in a divide by zero oops on ack generation.

Address the issue by checking the subflow status before
sending the ack.

Additionally protect mptcp_recvmsg() against invocation
with weird socket states.

v1 -> v2:
 - removed unneeded inline keyword - Jakub
Reported-and-suggested-by: Eric Dumazet <eric.dumazet@gmail.com>
Fixes: ea4ca586 ("mptcp: refine MPTCP-level ack scheduling")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Link: https://lore.kernel.org/r/5370c0ae03449239e3d1674ddcfb090cf6f20abe.1606253206.git.pabeni@redhat.comSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Extend MRP to support LC mode(link check) for the interconnect port.
This applies only to the interconnect ring.

Opposite to RC mode(ring check) the LC mode is using CFM frames to
detect when the link goes up or down and based on that the userspace
will need to react.
One advantage of the LC mode over RC mode is that there will be fewer
frames in the normal rings. Because RC mode generates InTest on all
ports while LC mode sends CFM frame only on the interconnect port.

All 4 nodes part of the interconnect ring needs to have the same mode.
And it is not possible to have running LC and RC mode at the same time
on a node.

Whenever the MIM starts it needs to detect the status of the other 3
nodes in the interconnect ring so it would send a frame called
InLinkStatus, on which the clients needs to reply with their link
status.

This patch adds InLinkStatus frame type and extends existing rules on
how to forward this frame.
Acked-by: Nikolay Aleksandrov <nikolay@nvidia.com>
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Link: https://lore.kernel.org/r/20201124082525.273820-1-horatiu.vultur@microchip.comSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Currently both filter and action flags use same "TCA_" prefix which makes
them hard to distinguish to code and confusing for users. Create aliases
for existing action flags constants with "TCA_ACT_" prefix.
Signed-off-by: Vlad Buslov <vlad@buslov.dev>
Link: https://lore.kernel.org/r/20201124164054.893168-1-vlad@buslov.devSigned-off-by: Jakub Kicinski <kuba@kernel.org>

On close this timer might be scheduled. mptcp uses sk_reset_timer for
this, so the a reference on the mptcp socket is taken.

This causes a refcount leak which can for example be reproduced
with 'mp_join_server_v4.pkt' from the mptcp-packetdrill repo.

The leak has nothing to do with join requests, v1_mp_capable_bind_no_cs.pkt
works too when replacing the last ack mpcapable to v1 instead of v0.

unreferenced object 0xffff888109bba040 (size 2744):
  comm "packetdrill", [..]
  backtrace:
    [..] sk_prot_alloc.isra.0+0x2b/0xc0
    [..] sk_clone_lock+0x2f/0x740
    [..] mptcp_sk_clone+0x33/0x1a0
    [..] subflow_syn_recv_sock+0x2b1/0x690 [..]

Fixes: e16163b6 ("mptcp: refactor shutdown and close")
Cc: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Link: https://lore.kernel.org/r/20201124162446.11448-1-fw@strlen.deSigned-off-by: Jakub Kicinski <kuba@kernel.org>

The rtl8211f supports downshift and before commit 5502b218
("net: phy: use phy_resolve_aneg_linkmode in genphy_read_status")
the read-back of register MII_CTRL1000 was used to detect the
negotiated link speed.
The code added in commit d445dff2 ("net: phy: realtek: read
actual speed to detect downshift") is working fine also for this
phy and it's trivial re-using it to restore the downshift
detection on rtl8211f.

Add the phy specific read_status() pointing to the existing
function rtlgen_read_status().
Signed-off-by: Antonio Borneo <antonio.borneo@st.com>
Link: https://lore.kernel.org/r/478f871a-583d-01f1-9cc5-2eea56d8c2a7@huawei.comTested-by: Yonglong Liu <liuyonglong@huawei.com>
Link: https://lore.kernel.org/r/20201124230756.887925-1-antonio.borneo@st.comSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Christian Eggers says:

====================
net: ptp: use common defines for PTP message types in further drivers

This series replaces further driver internal enumeration / uses of magic
numbers with the newly introduced PTP_MSGTYPE_* defines.
====================

Link: https://lore.kernel.org/r/20201124074418.2609-1-ceggers@arri.deSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Use recently introduced PTP_MSGTYPE_SYNC and PTP_MSGTYPE_DELAY_REQ
defines instead of a driver internal enumeration.
Signed-off-by: Christian Eggers <ceggers@arri.de>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Use recently introduced PTP wide defines instead of a driver internal
enumeration.
Signed-off-by: Christian Eggers <ceggers@arri.de>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Cc: Petr Machata <petrm@mellanox.com>
Cc: Jiri Pirko <jiri@nvidia.com>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Replace use of magic number with recently introduced define.
Signed-off-by: Christian Eggers <ceggers@arri.de>
Reviewed-by: Kurt Kanzenbach <kurt@linutronix.de>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Ioana Ciornei says:

====================
net: phy: add support for shared interrupts (part 3)

This patch set aims to actually add support for shared interrupts in
phylib and not only for multi-PHY devices. While we are at it,
streamline the interrupt handling in phylib.

For a bit of context, at the moment, there are multiple phy_driver ops
that deal with this subject:

- .config_intr() - Enable/disable the interrupt line.

- .ack_interrupt() - Should quiesce any interrupts that may have been
  fired.  It's also used by phylib in conjunction with .config_intr() to
  clear any pending interrupts after the line was disabled, and before
  it is going to be enabled.

- .did_interrupt() - Intended for multi-PHY devices with a shared IRQ
  line and used by phylib to discern which PHY from the package was the
  one that actually fired the interrupt.

- .handle_interrupt() - Completely overrides the default interrupt
  handling logic from phylib. The PHY driver is responsible for checking
  if any interrupt was fired by the respective PHY and choose
  accordingly if it's the one that should trigger the link state machine.

From my point of view, the interrupt handling in phylib has become
somewhat confusing with all these callbacks that actually read the same
PHY register - the interrupt status.  A more streamlined approach would
be to just move the responsibility to write an interrupt handler to the
driver (as any other device driver does) and make .handle_interrupt()
the only way to deal with interrupts.

Another advantage with this approach would be that phylib would gain
support for shared IRQs between different PHY (not just multi-PHY
devices), something which at the moment would require extending every
PHY driver anyway in order to implement their .did_interrupt() callback
and duplicate the same logic as in .ack_interrupt(). The disadvantage
of making .did_interrupt() mandatory would be that we are slightly
changing the semantics of the phylib API and that would increase
confusion instead of reducing it.

What I am proposing is the following:

- As a first step, make the .ack_interrupt() callback optional so that
  we do not break any PHY driver amid the transition.

- Every PHY driver gains a .handle_interrupt() implementation that, for
  the most part, would look like below:

	irq_status = phy_read(phydev, INTR_STATUS);
	if (irq_status < 0) {
		phy_error(phydev);
		return IRQ_NONE;
	}

	if (!(irq_status & irq_mask))
		return IRQ_NONE;

	phy_trigger_machine(phydev);

	return IRQ_HANDLED;

- Remove each PHY driver's implementation of the .ack_interrupt() by
  actually taking care of quiescing any pending interrupts before
  enabling/after disabling the interrupt line.

- Finally, after all drivers have been ported, remove the
  .ack_interrupt() and .did_interrupt() callbacks from phy_driver.

This patch set is part 3 (and final) of the entire change set and it
addresses the remaining PHY drivers that have not been migrated
previosly. Also, it finally removed the .did_interrupt() and
.ack_interrupt() callbacks since they are of no use anymore.

I do not have access to most of these PHY's, therefore I Cc-ed the
latest contributors to the individual PHY drivers in order to have
access, hopefully, to more regression testing.
====================

Link: https://lore.kernel.org/r/20201123153817.1616814-1-ciorneiioana@gmail.comSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Now that all the PHY drivers have been migrated to directly implement
the generic .handle_interrupt() callback for a seamless support of
shared IRQs and all the .config_inter() implementations clear any
pending interrupts, we can safely remove the two callbacks.

With this patch, phylib has a proper support for shared IRQs (and not
just for multi-PHY devices. A PHY driver must implement both the
.handle_interrupt() and .config_intr() callbacks for the IRQs to be
actually used.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

In preparation of removing the .ack_interrupt() callback, we must replace
its occurrences (aka phy_clear_interrupt), from the 2 places where it is
called from (phy_enable_interrupts and phy_disable_interrupts), with
equivalent functionality.

This means that clearing interrupts now becomes something that the PHY
driver is responsible of doing, before enabling interrupts and after
clearing them. Make this driver follow the new contract.

Also, add a comment describing the multiple step interrupt
acknoledgement process.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

In an attempt to actually support shared IRQs in phylib, we now move the
responsibility of triggering the phylib state machine or just returning
IRQ_NONE, based on the IRQ status register, to the PHY driver. Having
3 different IRQ handling callbacks (.handle_interrupt(),
.did_interrupt() and .ack_interrupt() ) is confusing so let the PHY
driver implement directly an IRQ handler like any other device driver.
Make this driver follow the new convention.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

In preparation of removing the .ack_interrupt() callback, we must replace
its occurrences (aka phy_clear_interrupt), from the 2 places where it is
called from (phy_enable_interrupts and phy_disable_interrupts), with
equivalent functionality.

This means that clearing interrupts now becomes something that the PHY
driver is responsible of doing, before enabling interrupts and after
clearing them. Make this driver follow the new contract.

Cc: Dan Murphy <dmurphy@ti.com>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

In an attempt to actually support shared IRQs in phylib, we now move the
responsibility of triggering the phylib state machine or just returning
IRQ_NONE, based on the IRQ status register, to the PHY driver. Having
3 different IRQ handling callbacks (.handle_interrupt(),
.did_interrupt() and .ack_interrupt() ) is confusing so let the PHY
driver implement directly an IRQ handler like any other device driver.
Make this driver follow the new convention.

Cc: Dan Murphy <dmurphy@ti.com>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

In preparation of removing the .ack_interrupt() callback, we must replace
its occurrences (aka phy_clear_interrupt), from the 2 places where it is
called from (phy_enable_interrupts and phy_disable_interrupts), with
equivalent functionality.

This means that clearing interrupts now becomes something that the PHY
driver is responsible of doing, before enabling interrupts and after
clearing them. Make this driver follow the new contract.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

In an attempt to actually support shared IRQs in phylib, we now move the
responsibility of triggering the phylib state machine or just returning
IRQ_NONE, based on the IRQ status register, to the PHY driver. Having
3 different IRQ handling callbacks (.handle_interrupt(),
.did_interrupt() and .ack_interrupt() ) is confusing so let the PHY
driver implement directly an IRQ handler like any other device driver.
Make this driver follow the new convention.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

In preparation of removing the .ack_interrupt() callback, we must replace
its occurrences (aka phy_clear_interrupt), from the 2 places where it is
called from (phy_enable_interrupts and phy_disable_interrupts), with
equivalent functionality.

This means that clearing interrupts now becomes something that the PHY
driver is responsible of doing, before enabling interrupts and after
clearing them. Make this driver follow the new contract.

Cc: Divya Koppera <Divya.Koppera@microchip.com>
Cc: Oleksij Rempel <o.rempel@pengutronix.de>
Cc: Philippe Schenker <philippe.schenker@toradex.com>
Cc: Marek Vasut <marex@denx.de>
Cc: Antoine Tenart <atenart@kernel.org>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

In an attempt to actually support shared IRQs in phylib, we now move the
responsibility of triggering the phylib state machine or just returning
IRQ_NONE, based on the IRQ status register, to the PHY driver. Having
3 different IRQ handling callbacks (.handle_interrupt(),
.did_interrupt() and .ack_interrupt() ) is confusing so let the PHY
driver implement directly an IRQ handler like any other device driver.
Make this driver follow the new convention.

Cc: Divya Koppera <Divya.Koppera@microchip.com>
Cc: Oleksij Rempel <o.rempel@pengutronix.de>
Cc: Philippe Schenker <philippe.schenker@toradex.com>
Cc: Marek Vasut <marex@denx.de>
Cc: Antoine Tenart <atenart@kernel.org>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

In preparation of removing the .ack_interrupt() callback, we must replace
its occurrences (aka phy_clear_interrupt), from the 2 places where it is
called from (phy_enable_interrupts and phy_disable_interrupts), with
equivalent functionality.

This means that clearing interrupts now becomes something that the PHY
driver is responsible of doing, before enabling interrupts and after
clearing them. Make this driver follow the new contract.

Cc: Jerome Brunet <jbrunet@baylibre.com>
Cc: Neil Armstrong <narmstrong@baylibre.com>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

In an attempt to actually support shared IRQs in phylib, we now move the
responsibility of triggering the phylib state machine or just returning
IRQ_NONE, based on the IRQ status register, to the PHY driver. Having
3 different IRQ handling callbacks (.handle_interrupt(),
.did_interrupt() and .ack_interrupt() ) is confusing so let the PHY
driver implement directly an IRQ handler like any other device driver.
Make this driver follow the new convention.

Cc: Jerome Brunet <jbrunet@baylibre.com>
Cc: Neil Armstrong <narmstrong@baylibre.com>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

In preparation of removing the .ack_interrupt() callback, we must replace
its occurrences (aka phy_clear_interrupt), from the 2 places where it is
called from (phy_enable_interrupts and phy_disable_interrupts), with
equivalent functionality.

This means that clearing interrupts now becomes something that the PHY
driver is responsible of doing, before enabling interrupts and after
clearing them. Make this driver follow the new contract.

Cc: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

In an attempt to actually support shared IRQs in phylib, we now move the
responsibility of triggering the phylib state machine or just returning
IRQ_NONE, based on the IRQ status register, to the PHY driver. Having
3 different IRQ handling callbacks (.handle_interrupt(),
.did_interrupt() and .ack_interrupt() ) is confusing so let the PHY
driver implement directly an IRQ handler like any other device driver.
Make this driver follow the new convention.

Cc: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

In preparation of removing the .ack_interrupt() callback, we must replace
its occurrences (aka phy_clear_interrupt), from the 2 places where it is
called from (phy_enable_interrupts and phy_disable_interrupts), with
equivalent functionality.

This means that clearing interrupts now becomes something that the PHY
driver is responsible of doing, before enabling interrupts and after
clearing them. Make this driver follow the new contract.

Cc: Mathias Kresin <dev@kresin.me>
Cc: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

In an attempt to actually support shared IRQs in phylib, we now move the
responsibility of triggering the phylib state machine or just returning
IRQ_NONE, based on the IRQ status register, to the PHY driver. Having
3 different IRQ handling callbacks (.handle_interrupt(),
.did_interrupt() and .ack_interrupt() ) is confusing so let the PHY
driver implement directly an IRQ handler like any other device driver.
Make this driver follow the new convention.

Cc: Mathias Kresin <dev@kresin.me>
Cc: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

The driver's ISR sends a 'software interrupt' event to the probe()
thread using the following method:
- probe(): write 0 to flag, enable s/w interrupt
- probe(): poll on flag, relax using usleep_range()
- ISR    : write 1 to flag

Replace with wake_up() / wait_event_timeout(). Besides being easier
to get right, this abstraction has better timing and memory
consistency properties.

Tested-by: Sven Van Asbroeck <thesven73@gmail.com> # lan7430
Signed-off-by: Sven Van Asbroeck <thesven73@gmail.com>
Link: https://lore.kernel.org/r/20201123191529.14908-2-TheSven73@gmail.comSigned-off-by: Jakub Kicinski <kuba@kernel.org>

For no apparent reason, this function reads the INT_STS register, and
checks if the software interrupt bit is set. These things have already
been carried out by this function's only caller.

Clean up by removing the redundant code.

Tested-by: Sven Van Asbroeck <thesven73@gmail.com> # lan7430
Signed-off-by: Sven Van Asbroeck <thesven73@gmail.com>
Link: https://lore.kernel.org/r/20201123191529.14908-1-TheSven73@gmail.comSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Lorenzo Bianconi says:

====================
mvneta: access skb_shared_info only on last frag

Build skb_shared_info on mvneta_rx_swbm stack and sync it to xdp_buff
skb_shared_info area only on the last fragment.
Avoid avoid unnecessary xdp_buff initialization in mvneta_rx_swbm routine.
This a preliminary series to complete xdp multi-buff in mvneta driver.
====================

Link: https://lore.kernel.org/r/cover.1605889258.git.lorenzo@kernel.orgSigned-off-by: Jakub Kicinski <kuba@kernel.org>

Build skb_shared_info on mvneta_rx_swbm stack and sync it to xdp_buff
skb_shared_info area only on the last fragment. Leftover cache miss in
mvneta_swbm_rx_frame will be addressed introducing mb bit in
xdp_buff/xdp_frame struct
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Pass skb_shared_info pointer from mvneta_xdp_put_buff caller. This is a
preliminary patch to reduce accesses to skb_shared_info area and reduce
cache misses.
Remove napi parameter in mvneta_xdp_put_buff signature since it is always
run in NAPI context
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Explicitly initialize mandatory fields defining xdp_buff struct
in mvneta_rx_swbm routine
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Tx/Rx FIFO is a HW resource limited by total size, but shared
by all ports of same CP110 and impacting port-performance.
Do not divide the FIFO for ports which are not enabled in DTS,
so active ports could have more FIFO.
No change in FIFO allocation if all 3 ports on the communication
processor enabled in DTS.

The active port mapping should be done in probe before FIFO-init.
Signed-off-by: Stefan Chulski <stefanc@marvell.com>
Reviewed-by: Russell King <rmk+kernel@armlinux.org.uk>
Link: https://lore.kernel.org/r/1606154073-28267-1-git-send-email-stefanc@marvell.comSigned-off-by: Jakub Kicinski <kuba@kernel.org>