For unicast transmission, the current NACK sending althorithm is over-
active that forces the sending side to retransmit a packet that is not
really lost but just arrived at the receiving side with some delay, or
even retransmit same packets that have already been retransmitted
before. As a result, many duplicates are observed also under normal
condition, ie. without packet loss.

One example case is: node1 transmits 1 2 3 4 10 5 6 7 8 9, when node2
receives packet #10, it puts into the deferdq. When the packet #5 comes
it sends NACK with gap [6 - 9]. However, shortly after that, when
packet #6 arrives, it pulls out packet #10 from the deferfq, but it is
still out of order, so it makes another NACK with gap [7 - 9] and so on
... Finally, node1 has to retransmit the packets 5 6 7 8 9 a number of
times, but in fact all the packets are not lost at all, so duplicates!

This commit reduces duplicates by changing the condition to send NACK,
also restricting the retransmissions on individual packets via a timer
of about 1ms. However, it also needs to say that too tricky condition
for NACKs or too long timeout value for retransmissions will result in
performance reducing! The criterias in this commit are found to be
effective for both the requirements to reduce duplicates but not affect
performance.

The tipc_link_rcv() is also improved to only dequeue skb from the link
deferdq if it is expected (ie. its seqno <= rcv_nxt).
Acked-by: NYing Xue <ying.xue@windriver.com>
Acked-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NTuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

During unicast link transmission, it's observed very often that because
of one or a few lost/dis-ordered packets, the sending side will fastly
reach the send window limit and must wait for the packets to be arrived
at the receiving side or in the worst case, a retransmission must be
done first. The sending side cannot release a lot of subsequent packets
in its transmq even though all of them might have already been received
by the receiving side.
That is, one or two packets dis-ordered/lost and dozens of packets have
to wait, this obviously reduces the overall throughput!

This commit introduces an algorithm to overcome this by using "Gap ACK
blocks". Basically, a Gap ACK block will consist of <ack, gap> numbers
that describes the link deferdq where packets have been got by the
receiving side but with gaps, for example:

      link deferdq: [1 2 3 4      10 11      13 14 15       20]
--> Gap ACK blocks:       <4, 5>,   <11, 1>,      <15, 4>, <20, 0>

The Gap ACK blocks will be sent to the sending side along with the
traditional ACK or NACK message. Immediately when receiving the message
the sending side will now not only release from its transmq the packets
ack-ed by the ACK but also by the Gap ACK blocks! So, more packets can
be enqueued and transmitted.
In addition, the sending side can now do "multi-retransmissions"
according to the Gaps reported in the Gap ACK blocks.

The new algorithm as verified helps greatly improve the TIPC throughput
especially under packet loss condition.

So far, a maximum of 32 blocks is quite enough without any "Too few Gap
ACK blocks" reports with a 5.0% packet loss rate, however this number
can be increased in the furture if needed.

Also, the patch is backward compatible.
Acked-by: NYing Xue <ying.xue@windriver.com>
Acked-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NTuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Heiner Kallweit says:

====================
net: phy: use generic PHY ability readers if callback get_features isn't set

Meanwhile we have generic functions for reading the abilities of
Clause 22 / 45 PHY's. This allows to use them as fallback in case
callback get_features isn't set. Benefit is the reduction of
boilerplate code in PHY drivers.

v2:
- adjust a comment in patch 1 to match the code
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now that phylib uses genphy_read_abilities() as fallback, we don't have
to set callback get_features any longer.
Signed-off-by: NHeiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: NAndrew Lunn <andrew@lunn.ch>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Meanwhile we have generic functions for reading the abilities of
Clause 22 / 45 PHY's. This allows to use them as fallback in case
callback get_features isn't set. Benefit is the reduction of
boilerplate code in PHY drivers.

v2:
- adjust the comment in phy_driver_register to match the code
Signed-off-by: NHeiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: NAndrew Lunn <andrew@lunn.ch>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Since the mcast conversion to rhashtable this function has been unused, so
remove it.
Signed-off-by: NNikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This reverts commit 6578229d.
netif_receive_skb_list() doesn't support GRO, therefore we may have
scenarios with decreased performance. See discussion here [0].

[0] https://marc.info/?t=155403847400001&r=1&w=2Signed-off-by: NHeiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Linux currently disable ECN for incoming connections when the SYN
requests ECN and the IP header has ECT(0)/ECT(1) set, as some
networks were reportedly mangling the ToS byte, hence could later
trigger false congestion notifications.

RFC8311 §4.3 relaxes RFC3168's requirements such that ECT can be set
one TCP control packets (including SYNs). The main benefit of this
is the decreased probability of losing a SYN in a congested
ECN-capable network (i.e., it avoids the initial 1s timeout).
Additionally, this allows the development of newer TCP extensions,
such as AccECN.

This patch relaxes the previous check, by enabling ECN on incoming
connections using SYN+ECT if at least one bit of the reserved flags
of the TCP header is set. Such bit would indicate that the sender of
the SYN is using a newer TCP feature than what the host implements,
such as AccECN, and is thus implementing RFC8311. This enables
end-hosts not supporting such extensions to still negociate ECN, and
to have some of the benefits of using ECN on control packets.
Signed-off-by: NOlivier Tilmans <olivier.tilmans@nokia-bell-labs.com>
Suggested-by: NBob Briscoe <research@bobbriscoe.net>
Cc: Koen De Schepper <koen.de_schepper@nokia-bell-labs.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NNeal Cardwell <ncardwell@google.com>
Acked-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Jiri Pirko says:

====================
net: extend devlink port attrs with switch ID

Extend devlink port attrs to contain switch ID and change drivers that
register devlink ports to use that.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently if the driver registers devlink port instance, he should set
the devlink port attributes as well. Then the devlink core is able to
obtain switch id itself, no need for driver to implement the ndo.
Once all drivers will implement devlink port registration, this ndo
should be removed. This warning guides new drivers to do things as
they should be done.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Pass the switch ID down the to devlink through devlink_port_attrs_set()
so it can be used by devlink_compat_switch_id_get(). Leave
ndo_get_port_parent_id implementation only for legacy.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Reviewed-by: NFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Obtain HW id and pass it down to mlxsw_core_port_init() as it would be
used as switch_id in devlink and exposed to user.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Remove implementation of get_port_parent_id ndo and rely on core calling
into devlink for the information directly.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Reviewed-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Pass the switch ID down the to devlink through devlink_port_attrs_set()
so it can be used by devlink_compat_switch_id_get().
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Reviewed-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Remove implementation of get_port_parent_id ndo and rely on core calling
into devlink for the information directly.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Pass the switch ID down the to devlink through devlink_port_attrs_set()
so it can be used by devlink_compat_switch_id_get().
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently the switch_id is being only initialized when switching eswitch
mode from "legacy" to "switchdev". However, nothing prevents the id to
be initialized from the very beginning. Physical ports can show it even
in "legacy" mode.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Remove implementation of get_port_parent_id ndo and rely on core calling
into devlink for the information directly.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Pass the switch ID down the to devlink through devlink_port_attrs_set()
so it can be used by devlink_compat_switch_id_get().
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Introduce devlink_compat_switch_id_get() helper which fills up switch_id
according to passed netdev pointer. Call it directly from
dev_get_port_parent_id() as a fallback when ndo_get_port_parent_id
is not defined for given netdev.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Extend devlink_port_attrs_set() to pass switch ID for ports which are
part of switch and store it in port attrs. For other ports, this is
NULL.

Note that this allows the driver to group devlink ports into one or more
switches according to the actual topology.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In order to save space in the struct, convert bools to bits.
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Reviewed-by: NFlorian Fainelli <f.fainelli@gmail.com>
Reviewed-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We can optimize the fdb convergence when a backup_port is present by not
immediately flushing the entries of the stopped port since traffic for
those entries will flow towards the backup_port.

There are 2 cases specifically that benefit most:
- when the stopped port comes up before the entries expire by themselves
- when there's an external entry refresh and they're kept while the
  backup_port is operating (e.g. mlag)
Signed-off-by: NNikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This header is not in use - remove it.
Signed-off-by: NYuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/net/ethernet/ibm/ibmvnic.c: In function '__ibmvnic_reset':
drivers/net/ethernet/ibm/ibmvnic.c:1971:21: warning: variable 'netdev' set but not used [-Wunused-but-set-variable]

It's never used since introduction in
commit ed651a10 ("ibmvnic: Updated reset handling")
Signed-off-by: NYueHaibing <yuehaibing@huawei.com>
Reviewed-by: NMukesh Ojha <mojha@codeaurora.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/net/ethernet/ibm/ehea/ehea_qmr.c: In function 'ehea_create_cq':
drivers/net/ethernet/ibm/ehea/ehea_qmr.c:127:7: warning: variable 'cq_handle_ref' set but not used [-Wunused-but-set-variable]
drivers/net/ethernet/ibm/ehea/ehea_qmr.c:126:15: warning: variable 'epa' set but not used [-Wunused-but-set-variable]

They are never used since commit
7a291083 ("[PATCH] ehea: IBM eHEA Ethernet Device Driver")
Signed-off-by: NYueHaibing <yuehaibing@huawei.com>
Reviewed-by: NMukesh Ojha <mojha@codeaurora.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/net/ethernet/pasemi/pasemi_mac.c: In function 'pasemi_mac_queue_csdesc':
drivers/net/ethernet/pasemi/pasemi_mac.c:1358:29: warning: variable 'cpyhdr' set but not used [-Wunused-but-set-variable]

It's never used since commit 8d636d8b ("pasemi_mac: jumbo
frame support") and can be removed.
Signed-off-by: NYueHaibing <yuehaibing@huawei.com>
Reviewed-by: NMukesh Ojha <mojha@codeaurora.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

pr->tx_bytes should be assigned to tx_bytes other than
rx_bytes.
Reported-by: NHulk Robot <hulkci@huawei.com>
Fixes: ce45b873 ("ehea: Fixing statistics")
Signed-off-by: NYueHaibing <yuehaibing@huawei.com>
Reviewed-by: NMukesh Ojha <mojha@codeaurora.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

syzbot reports:
BUG: using __this_cpu_read() in preemptible code:
caller is dev_recursion_level include/linux/netdevice.h:3052 [inline]
 __this_cpu_preempt_check+0x246/0x270 lib/smp_processor_id.c:47
 dev_recursion_level include/linux/netdevice.h:3052 [inline]
 ip6_skb_dst_mtu include/net/ip6_route.h:245 [inline]

I erronously downgraded a this_cpu_read to __this_cpu_read when
moving dev_recursion_level() around.

Reported-by: syzbot+51471b4aae195285a4a3@syzkaller.appspotmail.com
Fixes: 97cdcf37 ("net: place xmit recursion in softnet data")
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Reviewed-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

skb somehow dequeued out of inputq before processing, it causes to
NULL pointer and kernel crashed.

Add checking skb valid before using.

Fixes: c55c8eda ("tipc: smooth change between replicast and broadcast")
Reported-by: NTuong Lien Tong <tuong.t.lien@dektech.com.au>
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NHoang Le <hoang.h.le@dektech.com.au>
Acked-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Recent changes to TC flower remove the requirement for rtnl lock when
accessing and modifying filters. Refcounts now ensure access and deletion
do not happen concurrently. However, the reoffload function which cycles
through all filters and replays them to registered hw drivers is not
protected.

Use the fl_get_next_filter() function to cycle the filters for reoffload
and ensure the ref taken by this function is put when done with each
filter.
Signed-off-by: NJohn Hurley <john.hurley@netronome.com>
Reviewed-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NVlad Buslov <vladbu@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Simplify this code by updating bridge multicast stats from
maybe_deliver().

Note that commit 6db6f0ea ("bridge: multicast to unicast"), in case
the port flag BR_MULTICAST_TO_UNICAST is set, never updates the previous
port pointer, therefore it is always going to be different from the
existing port in this deduplicated list iteration.
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>
Acked-by: NNikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Just like 46cfd725 ("net: use kfree_skb_list() helper in more places").
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>
Acked-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Huazhong Tan says:

====================
code optimizations & bugfixes for HNS3 driver

This patchset includes bugfixes and code optimizations for
the HNS3 ethernet controller driver.

Change log:
V1->V2: fixes comments from Sergei Shtylyov
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The function hnae3_match_n_instantiate() was called both by
initializing or uninitializing client. For uninitializing, the
return value was never used.

To make it more clear, this patch splits it to two functions,
hnae3_init_client_instance() and hnae3_uninit_client_instance().
Signed-off-by: NJian Shen <shenjian15@huawei.com>
Signed-off-by: NPeng Li <lipeng321@huawei.com>
Signed-off-by: NHuazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In the hns3_reset_notify_uninit_enet() HNS3_NIC_STATE_INITED flag
should be checked and cleared firstly.
Signed-off-by: NHuazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: NPeng Li <lipeng321@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When resetting, the changing of MTU is not allowed, so this patch
adds checking reset status in hns3_nic_change_mtu() to do that.
Signed-off-by: NHuazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: NPeng Li <lipeng321@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

hns3_get_stats() should check the resetting status firstly,
since the device will be reinitialized when resetting. If the
reset has not completed, the hns3_get_stats() may access
invalid memory.
Signed-off-by: NHuazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: NPeng Li <lipeng321@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

It used netdev->uc and netdev->mc list in function
hns3_recover_hw_addr() and hns3_remove_hw_addr().
We should add protect for them.

Fixes: f05e2109 ("net: hns3: Clear mac vlan table entries when unload driver or function reset")
Signed-off-by: NJian Shen <shenjian15@huawei.com>
Signed-off-by: NPeng Li <lipeng321@huawei.com>
Signed-off-by: NHuazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

csq is used as a ring buffer, the value of the desc will be replaced
in next use. This patch removes the unnecessary memset, and just
updates the next_to_clean.
Signed-off-by: NHuazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: NPeng Li <lipeng321@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>