David S. Miller [Mon, 4 Mar 2019 04:41:18 +0000 (20:41 -0800)]
Merge branch 'dpaa2-eth-add-XDP_REDIRECT-support'
Ioana Ciornei says:
====================
dpaa2-eth: add XDP_REDIRECT support
The first patch adds different software annotation types for Tx frames
depending on frame type while the second one actually adds support for basic
XDP_REDIRECT.
Changes in v2:
- add missing xdp_do_flush_map() call
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Ioana Radulescu [Fri, 1 Mar 2019 17:47:24 +0000 (17:47 +0000)]
dpaa2-eth: add XDP_REDIRECT support
Implement support for the XDP_REDIRECT action.
The redirected frame is transmitted and confirmed on the regular Tx/Tx
conf queues. Frame is marked with the "XDP" type in the software
annotation, since it requires special treatment.
We don't have good hardware support for TX batching, so the
XDP_XMIT_FLUSH flag doesn't make a difference for now; ndo_xdp_xmit
performs the actual Tx operation on the spot.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com> Signed-off-by: Ioana Radulescu <ruxandra.radulescu@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ioana Radulescu [Fri, 1 Mar 2019 17:47:23 +0000 (17:47 +0000)]
dpaa2-eth: Add software annotation types
We write different metadata information in the software annotation
area of Tx frames, depending on frame type. Make this more explicit
by introducing a type field and separate structures for single buffer
and scatter-gather frames.
Signed-off-by: Ioana Radulescu <ruxandra.radulescu@nxp.com> Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
sched: Patches from out-of-tree version of sch_cake
This series includes a couple of patches with updates from the out-of-tree
version of sch_cake. The first one is a fix to the fairness scheduling when
dual-mode fairness is enabled. The second patch is an additional feature flag
that allows using fwmark as a tin selector, as a convenience for people who want
to customise tin selection. The third patch is just a cleanup to the tin
selection logic.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
With more modes added the logic in cake_select_tin() was getting a bit
hairy, and it turns out we can actually simplify it quite a bit. This also
allows us to get rid of one of the two diffserv parsing functions, which
has the added benefit that already-zeroed DSCP fields won't get re-written.
Suggested-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
sch_cake: Permit use of connmarks as tin classifiers
Add flag 'FWMARK' to enable use of firewall connmarks as tin selector.
The connmark (skbuff->mark) needs to be in the range 1->tin_cnt ie.
for diffserv3 the mark needs to be 1->3.
Background
Typically CAKE uses DSCP as the basis for tin selection. DSCP values
are relatively easily changed as part of the egress path, usually with
iptables & the mangle table, ingress is more challenging. CAKE is often
used on the WAN interface of a residential gateway where passthrough of
DSCP from the ISP is either missing or set to unhelpful values thus use
of ingress DSCP values for tin selection isn't helpful in that
environment.
An approach to solving the ingress tin selection problem is to use
CAKE's understanding of tc filters. Naive tc filters could match on
source/destination port numbers and force tin selection that way, but
multiple filters don't scale particularly well as each filter must be
traversed whether it matches or not. e.g. a simple example to map 3
firewall marks to tins:
MAJOR=$( tc qdisc show dev $DEV | head -1 | awk '{print $3}' )
tc filter add dev $DEV parent $MAJOR protocol all handle 0x01 fw action skbedit priority ${MAJOR}1
tc filter add dev $DEV parent $MAJOR protocol all handle 0x02 fw action skbedit priority ${MAJOR}2
tc filter add dev $DEV parent $MAJOR protocol all handle 0x03 fw action skbedit priority ${MAJOR}3
Another option is to use eBPF cls_act with tc filters e.g.
MAJOR=$( tc qdisc show dev $DEV | head -1 | awk '{print $3}' )
tc filter add dev $DEV parent $MAJOR bpf da obj my-bpf-fwmark-to-class.o
This has the disadvantages of a) needing someone to write & maintain
the bpf program, b) a bpf toolchain to compile it and c) needing to
hardcode the major number in the bpf program so it matches the cake
instance (or forcing the cake instance to a particular major number)
since the major number cannot be passed to the bpf program via tc
command line.
As already hinted at by the previous examples, it would be helpful
to associate tins with something that survives the Internet path and
ideally allows tin selection on both egress and ingress. Netfilter's
conntrack permits setting an identifying mark on a connection which
can also be restored to an ingress packet with tc action connmark e.g.
tc filter add dev eth0 parent ffff: protocol all prio 10 u32 \
match u32 0 0 flowid 1:1 action connmark action mirred egress redirect dev ifb1
Since tc's connmark action has restored any connmark into skb->mark,
any of the previous solutions are based upon it and in one form or
another copy that mark to the skb->priority field where again CAKE
picks this up.
This change cuts out at least one of the (less intuitive &
non-scalable) middlemen and permit direct access to skb->mark.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
George Amanakis [Fri, 1 Mar 2019 15:04:05 +0000 (16:04 +0100)]
sch_cake: Make the dual modes fairer
CAKE host fairness does not work well with TCP flows in dual-srchost and
dual-dsthost setup. The reason is that ACKs generated by TCP flows are
classified as sparse flows, and affect flow isolation from other hosts. Fix
this by calculating host_load based only on the bulk flows a host
generates. In a hash collision the host_bulk_flow_count values must be
decremented on the old hosts and incremented on the new ones *if* the queue
is in the bulk set.
Reported-by: Pete Heist <peteheist@gmail.com> Signed-off-by: George Amanakis <gamanakis@gmail.com> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
Macb power management support for ZynqMP
This series adds support for macb suspend/resume with system power down.
In relation to the above, this series also updates mdio_read/write
function for PM and adds tsu clock management.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Harini Katakam [Fri, 1 Mar 2019 10:50:35 +0000 (16:20 +0530)]
net: macb: Add support for suspend/resume with full power down
When macb device is suspended and system is powered down, the clocks
are removed and hence macb should be closed gracefully and restored
upon resume. This patch does the same by switching off the net device,
suspending phy and performing necessary cleanup of interrupts and BDs.
Upon resume, all these are reinitialized again.
Reset of macb device is done only when GEM is not a wake device.
Even when gem is a wake device, tx queues can be stopped and ptp device
can be closed (tsu clock will be disabled in pm_runtime_suspend) as
wake event detection has no dependency on this.
Signed-off-by: Kedareswara rao Appana <appanad@xilinx.com> Signed-off-by: Harini Katakam <harini.katakam@xilinx.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Harini Katakam [Fri, 1 Mar 2019 10:50:34 +0000 (16:20 +0530)]
net: macb: Add pm runtime support
Add runtime pm functions and move clock handling there.
Add runtime PM calls to mdio functions to allow for active mdio bus.
Signed-off-by: Shubhrajyoti Datta <shubhrajyoti.datta@xilinx.com> Signed-off-by: Harini Katakam <harini.katakam@xilinx.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Harini Katakam [Fri, 1 Mar 2019 10:50:33 +0000 (16:20 +0530)]
net: macb: Support clock management for tsu_clk
TSU clock needs to be enabled/disabled as per support in devicetree
and it should also be controlled during suspend/resume (WOL has no
dependency on this clock).
Signed-off-by: Harini Katakam <harini.katakam@xilinx.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Harini Katakam [Fri, 1 Mar 2019 10:50:32 +0000 (16:20 +0530)]
net: macb: Check MDIO state before read/write and use timeouts
Replace the while loop in MDIO read/write functions with a timeout.
In addition, add a check for MDIO bus busy before initiating a new
operation as well to make sure there is no ongoing MDIO operation.
Signed-off-by: Shubhrajyoti Datta <shubhrajyoti.datta@xilinx.com> Signed-off-by: Sai Pavan Boddu <sai.pavan.boddu@xilinx.com> Signed-off-by: Harini Katakam <harini.katakam@xilinx.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
net: dsa: microchip: add KSZ9893 switch support
This series of patches is to modify the KSZ9477 DSA driver to support
running KSZ9893 switch.
The KSZ9893 switch is similar to KSZ9477 except the ingress tail tag has
1 byte instead of 2 bytes. The XMII register that governs the MAC
communication also has different register definitions.
v1
- Put KSZ9893 tagging in separate patch
- Remove other switch support
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Tristram Ha [Fri, 1 Mar 2019 03:57:24 +0000 (19:57 -0800)]
net: dsa: microchip: add KSZ9893 switch support
Add KSZ9893 switch support in KSZ9477 driver. This switch is similar to
KSZ9477 except the ingress tail tag has 1 byte instead of 2 bytes, so
KSZ9893 tagging will be used.
The XMII register that governs how the host port communicates with the
MAC also has different register definitions.
Signed-off-by: Tristram Ha <Tristram.Ha@microchip.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Tristram Ha [Fri, 1 Mar 2019 03:57:23 +0000 (19:57 -0800)]
net: dsa: add KSZ9893 switch tagging support
KSZ9893 switch is similar to KSZ9477 switch except the ingress tail tag
has 1 byte instead of 2 bytes. The size of the portmap is smaller and
so the override and lookup bits are also moved.
Signed-off-by: Tristram Ha <Tristram.Ha@microchip.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Tristram Ha [Fri, 1 Mar 2019 03:57:22 +0000 (19:57 -0800)]
dt-bindings: net: dsa: document additional Microchip KSZ9477 family switches
Document additional Microchip KSZ9477 family switches.
Show how KSZ8565 switch should be configured as the host port is port 7
instead of port 5.
Signed-off-by: Tristram Ha <Tristram.Ha@microchip.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Sat, 2 Mar 2019 22:04:20 +0000 (14:04 -0800)]
Merge tag 'mlx5-updates-2019-03-01' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux
Saeed Mahameed says:
====================
mlx5-updates-2019-03-01
This series adds multipath offload support and contains some small updates
to mlx5 driver.
Multipath offload support from Roi Dayan:
We are going to track SW multipath route and related nexthops and reflect
that as port affinity to the HW.
1) Some patches are preparation.
2) add the multipath mode and fib events handling.
3) add support to handle offload failure for net error, i.e.
port down.
4) Small updates to match the behavior of multipath
Two small updates from Eran Ben Elisha,
5) Make a function static
6) Update PCIe supported devices list.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
====================
Netfilter/IPVS updates for net-next
The following patchset contains Netfilter/IPVS updates for net-next:
1) Add .release_ops to properly unroll .select_ops, use it from nft_compat.
After this change, we can remove list of extensions too to simplify this
codebase.
2) Update amanda conntrack helper to support v3.4, from Florian Tham.
3) Get rid of the obsolete BUGPRINT macro in ebtables, from
Florian Westphal.
4) Merge IPv4 and IPv6 masquerading infrastructure into one single module.
From Florian Westphal.
5) Patchset to remove nf_nat_l3proto structure to get rid of
indirections, from Florian Westphal.
6) Skip unnecessary conntrack timeout updates in case the value is
still the same, also from Florian Westphal.
7) Remove unnecessary 'fall through' comments in empty switch cases,
from Li RongQing.
8) Fix lookup to fixed size hashtable sets on big endian with 32-bit keys.
9) Incorrect logic to deactivate path of fixed size hashtable sets,
element was being tested to self.
10) Remove nft_hash_key(), the bitmap set is always selected for 16-bit
keys.
11) Use boolean whenever possible in IPVS codebase, from Andrea Claudi.
12) Enter close state in conntrack if RST matches exact sequence number,
from Florian Westphal.
13) Initialize dst_cache in tunnel extension, from wenxu.
14) Pass protocol as u16 to xt_check_match and xt_check_target, from
Li RongQing.
15) SCTP header is granted to be in a linear area from IPVS NAT handler,
from Xin Long.
16) Don't steal packets coming from slave VRF device from the
ip_sabotage_in() path, from David Ahern.
17) Fix unsafe update of basechain stats, from Li RongQing.
18) Make sure CONNTRACK_LOCKS is power of 2 to let compiler optimize
modulo operation as bitwise AND, from Li RongQing.
19) Use device_attribute instead of internal definition in the IDLETIMER
target, from Sami Tolvanen.
20) Merge redir, masq and IPv4/IPv6 NAT chain types, from Florian Westphal.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Here's one more bluetooth-next pull request for the 5.1 kernel:
- Added support for MediaTek MT7663U and MT7668U UART devices
- Cleanups & fixes to the hci_qca driver
- Fixed wakeup pin behavior for QCA6174A controller
Please let me know if there are any issues pulling. Thanks.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Sean Wang [Sat, 2 Mar 2019 18:44:09 +0000 (02:44 +0800)]
Bluetooth: mediatek: add support for MediaTek MT7663U and MT7668U UART devices
This adds the support of enabling MT7663U and MT7668U Bluetooth function
running on the top of btmtkuart driver.
There are a few differences between MT766[3,8]U and MT7622 where
MT766[3,8]U are standalone devices based on UART transport while MT7622
bluetooth is a built-in device on MediaTek SoC communicating with the host
through BTIF serial transport. Thus, extra setup sequence is necessary
for these standalone devices such as remote regulator and reset control via
GPIO, baud rate changing handshake between the host and device and so on.
Signed-off-by: Sean Wang <sean.wang@mediatek.com> Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Bluetooth: hci_qca: Reduce delay after sending baudrate request for WCN3990
The current 300ms delay after a baudrate change is extremely long.
For WCN3990 it is sufficient to wait 10ms after the baudrate change
request has been sent over the wire.
Ido Schimmel [Fri, 1 Mar 2019 13:38:43 +0000 (13:38 +0000)]
net: ipv4: Fix NULL pointer dereference in route lookup
When calculating the multipath hash for input routes the flow info is
not available and therefore should not be used.
Fixes: 24ba14406c5c ("route: Add multipath_hash in flowi_common to make user-define hash") Signed-off-by: Ido Schimmel <idosch@mellanox.com> Cc: wenxu <wenxu@ucloud.cn> Acked-by: wenxu <wenxu@ucloud.cn> Acked-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Sat, 2 Mar 2019 07:23:35 +0000 (23:23 -0800)]
Merge branch 'net-mvpp2-fixes-and-improvements'
Antoine Tenart says:
====================
net: mvpp2: fixes and improvements
This series aims to improve the Marvell PPv2 driver and to fix various
issues we encountered while testing the ports in many different
configurations. The series is based on top of Russell PPv2 phylink
rework and improvement.
I'm not sending a v2 of the previous fixes series as half the patches
are not the same and lots of development happened in between.
While this series contains fixes, it's sent to net-next as it is based
on top of Russell patches that were merged into net-next. I'm also
aiming at net-next as the series reworks critical paths of the PPv2
driver, such as the reset handling of various blocks, to let more weeks
for users to tests and for possible fixes to be sent before it lands
into a stable kernel version.
The series is divided into three parts:
- Patches 1 to 3 are cosmetic changes, sent alongside the series, as I
saw these small issues while working on this.
- Patches 5 to 8 are fixing (or improving) individual issues that we
found while testing PPv2.1 and PPv2.2 ports while using various
interfaces.
Notable fixes are we support back RGMII interfaces (on both PPv2.1 and
PPv2.2), as their support was broken by previous patches. We also
reworked the RXQ computation as the RXQ assignment was not checking
the maximum number of RXQ available, and was broken for PPv2.1.
- As discussed in a previous fixes series, patches 9 to 15 rework the
way blocks are set in reset in the PPv2 engine (plus related changes).
There are four blocks we want to control the reset status: two MAC
(GMAC and XLG MAC) and two PCS (MPCS and XPCS). The XLG MAC is used
for 10G connexions and uses the MPCS or the XPCS depending on the mode
used (10GKR / XAUI / RXAUI) and the GMAC is used for the other modes.
The idea is to set all blocks in reset by default, and when not used,
and to de-assert the reset only when a block is used. There are four
cases to take in account:
1. Boot time: all four blocks should be put in reset, as we do not
know their initial state (configured by the firmware/bootloader).
2. Link up: only the blocks used by a given mode should be put out of
reset (eg. 10GKR uses the XLG MAC and the MPCS).
3. Mode reconfiguration: some ports may support mode reconfiguration,
and switching between the GMAC and the XLG MAC (or between the two
PCS). All blocks should be put in reset, and only the one used
should be put out of reset.
4. Link down: all four blocks are put in reset.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Antoine Tenart [Fri, 1 Mar 2019 10:52:17 +0000 (11:52 +0100)]
net: mvpp2: set the GMAC, XLG MAC, XPCS and MPCS in reset when a port is down
This patch adds calls in the stop() helper to ensure both MACs and
both PCS blocks are set in reset when the user manually sets a port
down. This is done so that we have the exact same block reset states at
boot time and when a port is set down.
Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Antoine Tenart [Fri, 1 Mar 2019 10:52:16 +0000 (11:52 +0100)]
net: mvpp2: set the XPCS and MPCS in reset when not used
This patch sets both the XPCS and MPCS blocks in reset when they aren't
used. This is done both at boot time and when reconfiguring a port mode.
The advantage now is that only the PCS used is set out of reset when the
port is configured (10GKR uses the MCPS while RXAUI uses the XPCS).
Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Antoine Tenart [Fri, 1 Mar 2019 10:52:15 +0000 (11:52 +0100)]
net: mvpp2: reset the MACs when reconfiguring a port
This patch makes sure both PPv2 MACs (GMAC + XLG MAC) are set in reset
while a port is reconfigured. This is done so that we make sure a MAC is
in a reset state when not used, as only one of the two will be set out
of reset after the port is configured properly.
Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Antoine Tenart [Fri, 1 Mar 2019 10:52:14 +0000 (11:52 +0100)]
net: mvpp2: rework the XLG MAC reset handling
This patch reworks the way the XLG MAC is set in reset: the XLG MAC is
set in reset at probe time and taken out of this state only when used.
The idea is to move forward a situation where only the blocks used are
taken out of reset. This also has the effect to handle the GMAC and the
XLG MAC in a similar way (the GMAC already is set in reset at boot
time).
Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Antoine Tenart [Fri, 1 Mar 2019 10:52:13 +0000 (11:52 +0100)]
net: mvpp2: force the XLG MAC link up or down when not using in-band
This patch force the XLG MAC link state in the phylink link_up() and
link_down() helpers when not using in-band auto-negotiation. This mimics
what's already done for the GMAC and follows what's advised in the
phylink documentation.
Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Antoine Tenart [Fri, 1 Mar 2019 10:52:12 +0000 (11:52 +0100)]
net: mvpp2: only update the XLG configuration when needed
This patch improves the XLG configuration function, to only update the
XLG configuration register when a change is needed. This helps not
writing over and over the same XLG configuration each time phylink
request the MAC to be configured. This mimics the GMAC configuration
function.
Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Antoine Tenart [Fri, 1 Mar 2019 10:52:11 +0000 (11:52 +0100)]
net: mvpp2: always disable both MACs when disabling a port
This patch modifies the port_disable() helper to always disable both the
GMAC and the XLG MAC when called. At boot time we do not know of a port
was enabled in the firmware/bootloader, and if so what mode was used
(hence which of the two MACs was used).
This also help in implementing a logic where all blocks are disabled
when not used, and only enabled regarding the current mode used on a
given port.
Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Antoine Tenart [Fri, 1 Mar 2019 10:52:10 +0000 (11:52 +0100)]
net: mvpp2: some AN fields require the link to be down when updated
The GMAC configuration helper modifies values in the auto-negotiation
register. Some of its values require the port to be forced down when
modifying their values. This patches fixes the check made on the bit to
be updated in this register, so that the port is forced down when
needed. This fix cases where some of those parameters were updated, but
not taken into account, such as when using RGMII interfaces.
Fixes: d14e078f23cc ("net: marvell: mvpp2: only reprogram what is necessary on mac_config") Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Antoine Tenart [Fri, 1 Mar 2019 10:52:09 +0000 (11:52 +0100)]
net: mvpp2: fix the computation of the RXQs
The patch fixes the computation of RXQs being used by the PPv2 driver,
which is set depending on the PPv2 engine version and the queue mode
used. There are three cases:
- PPv2.1: 1 RXQ per CPU.
- PPV2.2 with MVPP2_QDIST_MULTI_MODE: 1 RXQ per CPU.
- PPv2.2 with MVPP2_QDIST_SINGLE_MODE: 1 RXQ is shared between the CPUs.
The PPv2 engine supports a maximum of 32 queues per port. This patch
adds a check so that we do not overstep this maximum.
It appeared the calculation was broken for PPv2.1 engines since f8c6ba8424b0, as PPv2.1 ports ended up with a single RXQ while they
needed 4. This patch fixes it.
Fixes: f8c6ba8424b0 ("net: mvpp2: use only one rx queue per port per CPU") Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Antoine Tenart [Fri, 1 Mar 2019 10:52:08 +0000 (11:52 +0100)]
net: mvpp2: fix validate for PPv2.1
The Phylink validate function is the Marvell PPv2 driver makes a check
on the GoP id. This is valid an has to be done when using PPv2.2 engines
but makes no sense when using PPv2.1. The check done when using an RGMII
interface makes sure the GoP id is not 0, but this breaks PPv2.1. Fixes
it.
Fixes: 0fb628f0f250 ("net: mvpp2: fix phylink handling of invalid PHY modes") Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Antoine Tenart [Fri, 1 Mar 2019 10:52:07 +0000 (11:52 +0100)]
net: mvpp2: reconfiguring the port interface is PPv2.2 specific
This patch adds a check on the PPv2 version in-use not to reconfigure
the port mode when an interface is updated when using PPv2.1 as the
functions called are PPv2.2 specific.
Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Antoine Tenart [Fri, 1 Mar 2019 10:52:06 +0000 (11:52 +0100)]
net: mvpp2: a port can be disabled even if we use the link IRQ
We had a check in the mvpp2_mac_link_down() function (called by phylink)
to avoid disabling the port when link interrupts are used. It turned out
the interrupt can still be used with the port disabled. We can thus
remove this check.
Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Antoine Tenart [Fri, 1 Mar 2019 10:52:04 +0000 (11:52 +0100)]
net: mvpp2: update the port documentation regarding the GoP
The Marvell PPv2 port structure stores the GoP id of a given port. This
information is specific to PPv2.2, but cannot be used by PPv2.1. Update
its comment to denote this specificity.
Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
cxgb4vf: Call netif_carrier_off properly in pci_probe
netif_carrier_off() should be called only after register_netdev().
Signed-off-by: Arjun Vynipadath <arjun@chelsio.com> Signed-off-by: Vishal Kulkarni <vishal@chelsio.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Reverting force link up changes since this behaviour can be
achieved using VF link state feature.
Reverts:
commit 0913667ab3ad ("cxgb4vf: Forcefully link up virtual interfaces")
Signed-off-by: Arjun Vynipadath <arjun@chelsio.com> Signed-off-by: Vishal Kulkarni <vishal@chelsio.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Use ndo_set_vf_link_state to control the link states associated
with the virtual interfaces.
Signed-off-by: Arjun Vynipadath <arjun@chelsio.com> Signed-off-by: Vishal Kulkarni <vishal@chelsio.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Some of these macros were conflicting with global namespace,
hence prefixing them with CXGB4VF.
Signed-off-by: Arjun Vynipadath <arjun@chelsio.com> Signed-off-by: Vishal Kulkarni <vishal@chelsio.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: YueHaibing <yuehaibing@huawei.com> Acked-by: Sean Wang <sean.wang@mediatek.com> for mt7530 and mtk_eth_soc Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Thu, 28 Feb 2019 23:17:28 +0000 (15:17 -0800)]
net: support 64bit rates for getsockopt(SO_MAX_PACING_RATE)
For legacy applications using 32bit variable, SO_MAX_PACING_RATE
has to cap the returned value to 0xFFFFFFFF, meaning that
rates above 34.35 Gbit are capped.
This patch allows applications to read socket pacing rate
at full resolution, if they provide a 64bit variable to store it,
and the kernel is 64bit.
Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Lucas Bates [Thu, 28 Feb 2019 22:38:40 +0000 (17:38 -0500)]
tc-testing: Allow test cases to be skipped
By adding a check for an optional key/value pair to the test case
data, individual test cases may be skipped to prevent tdc from
aborting a test run due to setup or teardown failure.
If a test case is skipped, it will still appear in the results
output to allow for a consistent number of executed tests in each
run. However, the test will be marked as skipped.
This support for skipping extends to any plugins that may generate
additional results for each executed test.
Signed-off-by: Lucas Bates <lucasb@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
When IPv6 is compiled but disabled at runtime, geneve_sock_add returns
-EAFNOSUPPORT. For metadata based tunnels, this causes failure of the whole
operation of bringing up the tunnel.
Ignore failure of IPv6 socket creation for metadata based tunnels caused by
IPv6 not being available.
This is the same fix as what commit d074bf960044 ("vxlan: correctly handle
ipv6.disable module parameter") is doing for vxlan.
Note there's also commit c0a47e44c098 ("geneve: should not call rt6_lookup()
when ipv6 was disabled") which fixes a similar issue but for regular
tunnels, while this patch is needed for metadata based tunnels.
Signed-off-by: Jiri Benc <jbenc@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Sat, 2 Mar 2019 05:44:11 +0000 (21:44 -0800)]
Merge branch 'mlxsw-rehash-split'
Ido Schimmel says:
====================
mlxsw: spectrum_acl: Split rehash work into chunks
Jiri says:
When rehash happens on a vregion with many rules and they are being
migrated, it might take significant time to finish the job. During that
time vregion->lock is taken which prevents rules from being
added/deleted from the vregion.
Aim of this patchset is to allow to interrupt migration of rules during
rehash, reschedule and give chance for rules to be added/deleted. Then
continue migration in another execution of scheduled work.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Thu, 28 Feb 2019 06:59:26 +0000 (06:59 +0000)]
mlxsw: spectrum_acl: Remember where to continue rehash migration
Store pointer to vchunk where the migration was interrupted, as well as
ventry pointer to start from and to stop at (during rollback). This
saved pointers need to be forgotten in case of ventries list or vchunk
list changes, which is done by couple of "changed" helpers.
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Ido Schimmel <idosch@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Thu, 28 Feb 2019 06:59:25 +0000 (06:59 +0000)]
mlxsw: spectrum_acl: Allow to interrupt/continue rehash work
Currently, migration of vregions with many entries may take long time
during which insertions and removals of the rules are blocked
due to wait to acquire vregion->lock.
To overcome this, allow to interrupt and continue rehash work according
to the set credits - number of rules to migrate.
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Ido Schimmel <idosch@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Thu, 28 Feb 2019 06:59:24 +0000 (06:59 +0000)]
mlxsw: spectrum_acl: Do rollback as another call to mlxsw_sp_acl_tcam_vchunk_migrate_all()
In order to simplify the code and to prepare it for
interrupted/continued migration process, do the rollback in case of
migration error as another call to mlxsw_sp_acl_tcam_vchunk_migrate_all().
It can be understood as "migrate all back".
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Ido Schimmel <idosch@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Thu, 28 Feb 2019 06:59:24 +0000 (06:59 +0000)]
mlxsw: spectrum_acl: Put vchunk migrate start/end code into separate functions
In preparations of interrupt/continue of rehash work, put the code that
is done at the beginning/end of vchunk migrate function into separate
functions.
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Ido Schimmel <idosch@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Thu, 28 Feb 2019 06:59:19 +0000 (06:59 +0000)]
mlxsw: spectrum_acl: Push code start/end from mlxsw_sp_acl_tcam_vregion_migrate()
Push code from the beginning and end of function
mlxsw_sp_acl_tcam_vregion_migrate() into rehash_start()/end() functions.
Then all the things needed to be done before and after the actual
migration process will be grouped together.
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Ido Schimmel <idosch@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Check if the entry is already in a chunk where we want it to be. In that
case, skip migration. This is preparation for "per parts" migration
where this situation may occur.
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Ido Schimmel <idosch@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Heiner Kallweit [Thu, 28 Feb 2019 06:39:15 +0000 (07:39 +0100)]
net: dsa: mv88e6xxx: prevent interrupt storm caused by mv88e6390x_port_set_cmode
When debugging another issue I faced an interrupt storm in this
driver (88E6390, port 9 in SGMII mode), consisting of alternating
link-up / link-down interrupts. Analysis showed that the driver
wanted to set a cmode that was set already. But so far
mv88e6390x_port_set_cmode() doesn't check this and powers down
SERDES, what causes the link to break, and eventually results in
the described interrupt storm.
Fix this by checking whether the cmode actually changes. We want
that the very first call to mv88e6390x_port_set_cmode() always
configures the registers, therefore initialize port.cmode with
a value that is different from any supported cmode value.
We have to take care that we only init the ports cmode once
chip->info->num_ports is set.
v2:
- add small helper and init the number of actual ports only
Fixes: 364e9d7776a3 ("net: dsa: mv88e6xxx: Power on/off SERDES on cmode change") Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Fainelli [Thu, 28 Feb 2019 00:29:16 +0000 (16:29 -0800)]
switchdev: Remove unused transaction item queue
There are no more in tree users of the
switchdev_trans_item_{dequeue,enqueue} or switchdev_trans_item structure
in the kernel since commit 00fc0c51e35b ("rocker: Change world_ops API
and implementation to be switchdev independant").
Remove this unused code and update the documentation accordingly since.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann [Fri, 1 Mar 2019 21:05:29 +0000 (22:05 +0100)]
bpf: fix sanitation rewrite in case of non-pointers
Marek reported that he saw an issue with the below snippet in that
timing measurements where off when loaded as unpriv while results
were reasonable when loaded as privileged:
[...]
uint64_t a = bpf_ktime_get_ns();
uint64_t b = bpf_ktime_get_ns();
uint64_t delta = b - a;
if ((int64_t)delta > 0) {
[...]
Turns out there is a bug where a corner case is missing in the fix d3bd7413e0ca ("bpf: fix sanitation of alu op with pointer / scalar
type from different paths"), namely fixup_bpf_calls() only checks
whether aux has a non-zero alu_state, but it also needs to test for
the case of BPF_ALU_NON_POINTER since in both occasions we need to
skip the masking rewrite (as there is nothing to mask).
Fixes: d3bd7413e0ca ("bpf: fix sanitation of alu op with pointer / scalar type from different paths") Reported-by: Marek Majkowski <marek@cloudflare.com> Reported-by: Arthur Fabre <afabre@cloudflare.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/netdev/CAJPywTJqP34cK20iLM5YmUMz9KXQOdu1-+BZrGMAGgLuBWz7fg@mail.gmail.com/T/ Acked-by: Song Liu <songliubraving@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
====================
doc: net: ieee802154: move from plain text to rst
The ieee802154 subsystem doc was still in plain text. With the networking book
taking shape I thought it was time to do the first step and move it over to rst.
This really is only the minimal conversion. I need to take some time to update
and extend the docs.
The patches are based on net-next, but they only touch the networking book so I
would not expect and trouble. From what I have seen they would go through
Jonathan's tree after being acked by Dave? If you want this patches against a
different tree let me know.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Moving the ieee802154 docs from a plain text file into the new rst
style. This commit only does the minimal needed change to bring the
documentation over. Follow up patches will improve and extend on this.
Signed-off-by: Stefan Schmidt <stefan@datenfreihafen.org> Tested-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Wed, 27 Feb 2019 19:36:36 +0000 (11:36 -0800)]
devlink: fix kdoc
devlink suffers from a few kdoc warnings:
net/core/devlink.c:5292: warning: Function parameter or member 'dev' not described in 'devlink_register'
net/core/devlink.c:5351: warning: Function parameter or member 'port_index' not described in 'devlink_port_register'
net/core/devlink.c:5753: warning: Function parameter or member 'parent_resource_id' not described in 'devlink_resource_register'
net/core/devlink.c:5753: warning: Function parameter or member 'size_params' not described in 'devlink_resource_register'
net/core/devlink.c:5753: warning: Excess function parameter 'top_hierarchy' description in 'devlink_resource_register'
net/core/devlink.c:5753: warning: Excess function parameter 'reload_required' description in 'devlink_resource_register'
net/core/devlink.c:5753: warning: Excess function parameter 'parent_reosurce_id' description in 'devlink_resource_register'
net/core/devlink.c:6451: warning: Function parameter or member 'region' not described in 'devlink_region_snapshot_create'
net/core/devlink.c:6451: warning: Excess function parameter 'devlink_region' description in 'devlink_region_snapshot_create'
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Nikita Danilov [Wed, 27 Feb 2019 12:10:13 +0000 (12:10 +0000)]
net: aquantia: use better wrappers for state registers
Replace some direct registers reads with better
online functions.
Signed-off-by: Nikita Danilov <nikita.danilov@aquantia.com> Signed-off-by: Igor Russkikh <igor.russkikh@aquantia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Nikita Danilov [Wed, 27 Feb 2019 12:10:11 +0000 (12:10 +0000)]
net: aquantia: replace AQ_HW_WAIT_FOR with readx_poll_timeout_atomic
David noticed the original define was hiding 'err' variable
reference. Thats confusing and counterintuitive.
Andrew noted the whole macro could be replaced with standard readx_poll
kernel macro. This makes code more readable.
Signed-off-by: Nikita Danilov <nikita.danilov@aquantia.com> Signed-off-by: Igor Russkikh <igor.russkikh@aquantia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Igor Russkikh [Wed, 27 Feb 2019 12:10:09 +0000 (12:10 +0000)]
net: aquantia: fixed instack structure overflow
This is a real stack undercorruption found by kasan build.
The issue did no harm normally because it only overflowed
2 bytes after `bitary` array which on most architectures
were mapped into `err` local.
Fixes: bab6de8fd180 ("net: ethernet: aquantia: Atlantic A0 and B0 specific functions.") Signed-off-by: Nikita Danilov <nikita.danilov@aquantia.com> Signed-off-by: Igor Russkikh <igor.russkikh@aquantia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
In reality msix_entry_mask always restricts number of iterations.
Adding extra condition to make logic clear and smatch happy.
Signed-off-by: Nikita Danilov <nikita.danilov@aquantia.com> Signed-off-by: Igor Russkikh <igor.russkikh@aquantia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Nikita Danilov [Wed, 27 Feb 2019 12:10:05 +0000 (12:10 +0000)]
net: aquantia: added newline at end of file
drivers/net/ethernet/aquantia/atlantic/aq_nic.c: 991:1:
warning: no newline at end of file
Signed-off-by: Nikita Danilov <nikita.danilov@aquantia.com> Signed-off-by: Igor Russkikh <igor.russkikh@aquantia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Nikita Danilov [Wed, 27 Feb 2019 12:10:03 +0000 (12:10 +0000)]
net: aquantia: fixed memcpy size
Not careful array dereference caused analysis tools
to think there could be memory overflow.
There was actually no corruption because the array is
two dimensional.
drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c: 140
aq_ethtool_get_strings() error:
memcpy() '*aq_ethtool_stat_names' too small (32 vs 704)
Signed-off-by: Nikita Danilov <nikita.danilov@aquantia.com> Signed-off-by: Igor Russkikh <igor.russkikh@aquantia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Hangbin Liu [Wed, 27 Feb 2019 08:15:29 +0000 (16:15 +0800)]
ipv4: Add ICMPv6 support when parse route ipproto
For ip rules, we need to use 'ipproto ipv6-icmp' to match ICMPv6 headers.
But for ip -6 route, currently we only support tcp, udp and icmp.
Add ICMPv6 support so we can match ipv6-icmp rules for route lookup.
v2: As David Ahern and Sabrina Dubroca suggested, Add an argument to
rtm_getroute_parse_ip_proto() to handle ICMP/ICMPv6 with different family.
Reported-by: Jianlin Shi <jishi@redhat.com> Fixes: eacb9384a3fe ("ipv6: support sport, dport and ip_proto in RTM_GETROUTE") Signed-off-by: Hangbin Liu <liuhangbin@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Paul Burton [Fri, 1 Mar 2019 22:58:09 +0000 (22:58 +0000)]
MIPS: eBPF: Fix icache flush end address
The MIPS eBPF JIT calls flush_icache_range() in order to ensure the
icache observes the code that we just wrote. Unfortunately it gets the
end address calculation wrong due to some bad pointer arithmetic.
The struct jit_ctx target field is of type pointer to u32, and as such
adding one to it will increment the address being pointed to by 4 bytes.
Therefore in order to find the address of the end of the code we simply
need to add the number of 4 byte instructions emitted, but we mistakenly
add the number of instructions multiplied by 4. This results in the call
to flush_icache_range() operating on a memory region 4x larger than
intended, which is always wasteful and can cause crashes if we overrun
into an unmapped page.
Fix this by correcting the pointer arithmetic to remove the bogus
multiplication, and use braces to remove the need for a set of brackets
whilst also making it obvious that the target field is a pointer.
Signed-off-by: Paul Burton <paul.burton@mips.com> Fixes: b6bd53f9c4e8 ("MIPS: Add missing file for eBPF JIT.") Cc: Alexei Starovoitov <ast@kernel.org> Cc: Daniel Borkmann <daniel@iogearbox.net> Cc: Martin KaFai Lau <kafai@fb.com> Cc: Song Liu <songliubraving@fb.com> Cc: Yonghong Song <yhs@fb.com> Cc: netdev@vger.kernel.org Cc: bpf@vger.kernel.org Cc: linux-mips@vger.kernel.org Cc: stable@vger.kernel.org # v4.13+ Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Eran Ben Elisha [Sun, 27 Jan 2019 13:01:25 +0000 (15:01 +0200)]
net/mlx5: Update the list of the PCI supported devices
Add the upcoming ConnectX-6 Dx.
In addition, add "ConnectX Family mlx5Gen Virtual Function" device ID.
Every new HCA VF will be identified with this device ID. Different VF
models will be distinguished by their revision id.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Reviewed-by: Aya Levin <ayal@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Roi Dayan [Tue, 5 Feb 2019 07:41:43 +0000 (09:41 +0200)]
net/mlx5e: Update check for merged eswitch device
The current check only validates if both netdevs use the same ops
which means both are vf reps or both uplink reps.
Unlike the case where the two uplinks are bonded (VF LAG), under
multipath scheme the switchdev parent id is not unified between the
uplink reps (and all the associated vf reps). However, we still want
to duplicate in the driver encap flows, adjust the merged eswitch
check for that matter.
Signed-off-by: Roi Dayan <roid@mellanox.com> Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Roi Dayan [Tue, 5 Feb 2019 07:48:38 +0000 (09:48 +0200)]
net/mlx5e: Use hint to resolve route when in HW multipath mode
As part of creating the tunnel headers while offloading TC encap rules,
we resolve the route and neighbour in order to get the source /
destination mac.
Since the way we offload multipath route is by having two HW rules,
one per uplink port, doing naive route lookup might get us a "wrong"
routing path which goes through the peer uplink and this will get us
eventually to create a wrong L2 header for the tunnel.
To avoid that, we use a device hint to get the correct route.
Signed-off-by: Roi Dayan <roid@mellanox.com> Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Under multipath when encap rules are duplicated to HW in the driver,
it's possible for one flow to be currently un-offloaded (e.g. lack of
next-hop route or neigh entry) while the other flow is offloaded. As
such, we move to query the counters of both flows at all times.
Signed-off-by: Roi Dayan <roid@mellanox.com> Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Roi Dayan [Wed, 6 Feb 2019 13:43:51 +0000 (15:43 +0200)]
net/mlx5e: Re-attempt to offload flows on multipath port affinity events
Under multipath it's possible for us to offload the flow only through
the e-switch for which proper route through the uplink exists.
When the port is up and the next-hop route is set again we want to
offload through it as well.
We generate SW event from the FIB event handler when multipath port
affinity changes. The tc offloads code gets this event, goes over the
flows which were marked as of having missing route and attempts to
offload them.
Signed-off-by: Roi Dayan <roid@mellanox.com> Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Roi Dayan [Thu, 21 Feb 2019 14:29:27 +0000 (16:29 +0200)]
net/mlx5: Emit port affinity event for multipath offloads
Under multipath offload scheme, as part of handling fib events, emit
mlx5 port affinity event on the enabled ports which will be handled by
the tc offloads code.
Signed-off-by: Roi Dayan <roid@mellanox.com> Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Roi Dayan [Tue, 5 Feb 2019 15:59:58 +0000 (17:59 +0200)]
net/mlx5e: Allow one failure when offloading tc encap rules under multipath
In a similar manner to uplink/VF LAG, under multipath we add encap peer
rule on the second port as well.
However, unlike the LAG case, we do want to allow failure for adding
one of the rules. This happens due to using a routing hint while doing
the route lookup when one path (next hop device) is down.
Introduce a new flag to indicate that route lookup failed for encap
flow. Note that a flow may still not be offloaded to hw due to missing
neighbour, in that case, the neigh update event will take care of it.
Signed-off-by: Roi Dayan <roid@mellanox.com> Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Roi Dayan [Tue, 19 Feb 2019 14:38:56 +0000 (16:38 +0200)]
net/mlx5e: Don't inherit flow flags on peer flow creation
Currently the peer flow inherits the flags from the original flow
after we've set it. At this time the flags are set according to
the flow state, e.g marked as going to slow path and such.
Even if not getting us to real bugs now, this opens the door to
get us to troubles later. Future proof the code and avoid the
inheritance, use the peer flags as were set on input when we
started adding the original flow.
Signed-off-by: Roi Dayan <roid@mellanox.com> Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Roi Dayan [Sun, 17 Feb 2019 09:54:06 +0000 (11:54 +0200)]
net/mlx5e: Activate HW multipath and handle port affinity based on FIB events
To support multipath offload we are going to track SW multipath route
and related nexthops. To do that we register to FIB notifier and handle
the route and next-hops events and reflect that as port affinity to HW.
When there is a new multipath route entry that all next-hops are the
ports of an HCA we will activate LAG in HW.
Egress wise, we use HW LAG as the means to emulate multipath on current
HW which doesn't support port selection based on xmit hash. In the
presence of multiple VFs which use multiple SQs (send queues) this
yields fairly good distribution.
HA wise, HW LAG buys us the ability for a given RQ (receive queue) to
receive traffic from both ports and for SQs to migrate xmitting over
the active port if their base port fails.
When the route entry is being updated to single path we will update
the HW port affinity to use that port only.
If a next-hop becomes dead we update the HW port affinity to the living
port.
When all next-hops are alive again we reset the affinity to default.
Due to FW/HW limitations, when a route is deleted we are not disabling
the HW LAG since doing so will not allow us to enable it again while
VFs are bounded. Typically this is just a temporary state when a
routing daemon removes dead routes and later adds them back as needed.
This patch only handles events for AF_INET.
Signed-off-by: Roi Dayan <roid@mellanox.com> Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Roi Dayan [Thu, 21 Feb 2019 16:24:48 +0000 (18:24 +0200)]
net/mlx5: Add multipath mode
In order to offload ecmp-on-host scheme where next-hop routes are used,
we will make use of HW LAG. Add accessor function to let upper layers
in the driver to realize if the lag acts in multi-path mode.
Signed-off-by: Roi Dayan <roid@mellanox.com> Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
projects / mirror_ubuntu-kernels.git / log