This reverts commit 3cc49492.

There is nothing wrong with coalescing during defragmentation, it
reduces truesize overhead and simplifies things for the receiving
socket (no fraglist walk needed).

However, it also destroys geometry of the original fragments.
While that doesn't cause any breakage (we make sure to not exceed largest
original size) ip_do_fragment contains a 'fastpath' that takes advantage
of a present frag list and results in fragments that (in most cases)
match what was received.

In case its needed the coalescing could be done later, when we're sure
the skb is not forwarded.  But discussion during NFWS resulted in
'lets just remove this for now'.

Cc: Eric Dumazet <edumazet@google.com>
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Sathya Perla says:

====================
be2net: patch set

Hi David, the following patch set has code cleanup patches, minor enhancements
and non-critical fixes. Pls consider applying to the net-next tree. Thanks!

Patch 1 removes duplicate code in be_setup_wol() routine making it simpler
and more readable.

Patch 2 fixes the the bridge mode return value for the ndo_bridge_getlink()
call. Instead of just relying on the SRIOV enabled state, the driver now
queries the FW, for the actual mode of bridge.

Patch 3 removes code for setting D0 power state as it's already done
in pci_enable_device()

Patch 4 fixes a bad return value in be_check_ufi_compatibility() routine
introduced by an earlier commit.

Patch 5 fixes a field in udp header being accessed while in network endian
format.

Patch 6 fixes the be_mcc_notify() routine to return an error status when
the FW/HW is in an error state.

Patch 7 fixes the be_cmd_rx_filter() routine to issue the RX_FILTER cmd
and not wait for a completion from the FW. If the FW/adapter
is in an error state, this change helps in not holding up the rtnl_lock
and keeping bottom halves disabled while the driver timesout waiting for
a response from the FW.

Patch 8 fixes the be_cmd_set_loopback() routine to issue the LOOPBACK cmd
and not wait for the FW completion while spin_lock_bh() is held on the
mcc_lock. As the cmd is always issued from ethtool in a process context,
it can sleep till the FW completion is received.

Patch 9 bumps up the driver version to 10.6.0.3
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NSathya Perla <sathya.perla@avagotech.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The SET_LOOPBACK_MODE command is always issued from ethtool only in a
process context. So, while waiting for the cmd to complete, the driver
can sleep instead of holding spin_lock_bh() on the mcc_lock. This is done
by calling be_mcc_notify() instead of be_mcc_notify_wait() (that returns
only after the cmd completes while the MCCQ is locked).
Signed-off-by: NSuresh Reddy <suresh.reddy@avagotech.com>
Signed-off-by: NSathya Perla <sathya.perla@avagotech.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This fix makes the RX_FILTER cmd asynchronous, i.e., the caller issues
this cmd and doesn't wait for a completion from the FW. If the FW/adapter
is in an error state, this change helps in not holding up the rtnl_lock
and keeping bottom halves disabled while the driver timesout waiting for
a response from the FW.
Signed-off-by: NSuresh Reddy <suresh.reddy@avagotech.com>
Signed-off-by: NSathya Perla <sathya.perla@avagotech.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When the adapter is in error state, return error from be_mcc_notify()
so that the caller routines need not sleep waiting for a response.
Signed-off-by: NSuresh Reddy <suresh.reddy@avagotech.com>
Signed-off-by: NSathya Perla <sathya.perla@avagotech.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The "dest" field in the UDP-hdr of a TX skb is in network endian format.
Convert it to host endian before accessing it. The os2bmc patch,
mentioned below introduced this code.

Fixes: 760c295e ("be2net: Support for OS2BMC")
Signed-off-by: NVenkat Duvvuru <VenkatKumar.Duvvuru@Emulex.com>
Signed-off-by: NSathya Perla <sathya.perla@avagotech.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In the commit a6e6ff6e
("be2net: simplify UFI compatibility checking"), a return value of "-1"
was incorrectly used in place of "false". This patch fixes it.

Fixes: a6e6ff6e ("be2net: simplify UFI compatibility checking")
Signed-off-by: NVasundhara Volam <vasundhara.volam@avagotech.com>
Signed-off-by: NSathya Perla <sathya.perla@avagotech.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

pci_enable_device() call sets device power state to D0; there is no need
doing it again.
Signed-off-by: NKalesh AP <kalesh.purayil@avagotech.com>
Signed-off-by: NSathya Perla <sathya.perla@avagotech.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The current code assumes that bridge functionality (EVB) in the adapter
is enabled only when SR-IOV is enabled. This is not always true.
This patch uses the GET_HSW_CONFIG FW cmd to query this from the FW.
Signed-off-by: NKalesh AP <kalesh.purayil@avagotech.com>
Signed-off-by: NSathya Perla <sathya.perla@avagotech.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This change will make be_setup_wol() routine more compact and readable
by removing some duplicate code.
Signed-off-by: NKalesh AP <kalesh.purayil@avagotech.com>
Signed-off-by: NSathya Perla <sathya.perla@avagotech.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If outgoing interface is specified and the candidate address is
restricted to the outgoing interface, it is enough to iterate
over that given interface only.
Signed-off-by: NYOSHIFUJI Hideaki <hideaki.yoshifuji@miraclelinux.com>
Acked-by: NErik Kline <ek@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

platform_driver does not need to set an owner because
platform_driver_register() will set it.
Signed-off-by: NKrzysztof Kozlowski <k.kozlowski@samsung.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Support manually setting the polarity to mdi or mdix
Signed-off-by: NDavid Thomson <david.thomson@alliedtelesis.co.nz>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Pass the mdix setting from ethtool down to the phy driver, to allow
driver specific implementations of manually setting the polarity.
Signed-off-by: NDavid Thomson <david.thomson@alliedtelesis.co.nz>
Reviewed-by: NFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

After commit 900f65d3 ("tcp: move duplicate code from
tcp_v4_init_sock()/tcp_v6_init_sock()"), we no longer
need to export tcp_init_xmit_timers()
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Acked-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fill also the port group state when sending notifications.
Signed-off-by: NSatish Ashok <sashok@cumulusnetworks.com>
Signed-off-by: NNikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

flags local variable in __mkroute_input is not used as a variable.
Signed-off-by: NMasatake YAMATO <yamato@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add support to allow non-local binds similar to how this was done for IPv4.
Non-local binds are very useful in emulating the Internet in a box, etc.

This add the ip_nonlocal_bind sysctl under ipv6.

Testing:

Set up nonlocal binding and receive routing on a host, e.g.:

ip -6 rule add from ::/0 iif eth0 lookup 200
ip -6 route add local 2001:0:0:1::/64 dev lo proto kernel scope host table 200
sysctl -w net.ipv6.ip_nonlocal_bind=1

Set up routing to 2001:0:0:1::/64 on peer to go to first host

ping6 -I 2001:0:0:1::1 peer-address -- to verify
Signed-off-by: NTom Herbert <tom@herbertland.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Eric Dumazet says:

====================
inet: timewait cleanups

Another round of patches to make tw handling simpler.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

inet_twsk_deschedule() calls are followed by inet_twsk_put().

Only particular case is in inet_twsk_purge() but there is no point
to defer the inet_twsk_put() after re-enabling BH.

Lets rename inet_twsk_deschedule() to inet_twsk_deschedule_put()
and move the inet_twsk_put() inside.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

timewait sockets have a complex refcounting logic.
Once we realize it should be similar to established and
syn_recv sockets, we can use sk_nulls_del_node_init_rcu()
and remove inet_twsk_unhash()

In particular, deferred inet_twsk_put() added in commit
13475a30 ("tcp: connect() race with timewait reuse")
looks unecessary : When removing a timewait socket from
ehash or bhash, caller must own a reference on the socket
anyway.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Kernel will crash the same if one of the pointer is NULL anyway.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Hop was always either 0 or sizeof(struct ipv6hdr).
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

MV88E6320 and MV88E6321 are largely compatible to MV886352,
but are members of a different chip family.
Signed-off-by: NAleksey S. Kazantsev <ioctl@yandex.ru>
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>
Reviewed-by: NAndrew Lunn <andrew@lunn.ch>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Yuchung Cheng says:

====================
tcp: fixes some congestion control corner cases

This patch series fixes corner cases of TCP congestion control.
First issue is to avoid continuing slow start when cwnd reaches ssthresh.
Second issue is incorrectly processing order of congestion state and
cwnd update when entering fast recovery or undoing cwnd.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The congestion state and cwnd can be updated in the wrong order.
For example, upon receiving a dubious ACK, we incorrectly raise
the cwnd first (tcp_may_raise_cwnd()/tcp_cong_avoid()) because
the state is still Open, then enter recovery state to reduce cwnd.

For another example, if the ACK indicates spurious timeout or
retransmits, we first revert the cwnd reduction and congestion
state back to Open state.  But we don't raise the cwnd even though
the ACK does not indicate any congestion.

To fix this problem we should first call tcp_fastretrans_alert() to
process the dubious ACK and update the congestion state, then call
tcp_may_raise_cwnd() that raises cwnd based on the current state.
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NNandita Dukkipati <nanditad@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In the original design slow start is only used to raise cwnd
when cwnd is stricly below ssthresh. It makes little sense
to slow start when cwnd == ssthresh: especially
when hystart has set ssthresh in the initial ramp, or after
recovery when cwnd resets to ssthresh. Not doing so will
also help reduce the buffer bloat slightly.
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NNandita Dukkipati <nanditad@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add a helper to test the slow start condition in various congestion
control modules and other places. This is to prepare a slight improvement
in policy as to exactly when to slow start.
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NNandita Dukkipati <nanditad@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This change makes it so that the call skb_defer_rx_timestamp will first
check for a phydev before going in and manipulating the skb->data and
skb->len values.  By doing this we can avoid unnecessary work on network
devices that don't support phydev.  As a result we reduce the total
instruction count needed to process this on most devices.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

V1 of this patch contains Eric Dumazet's suggestion to move the per
dst RTAX_QUICKACK check into tcp_in_quickack_mode(). Thanks Eric.

I ran some tests and after setting the "ip route change quickack 1"
knob there were still many delayed ACKs sent. This occured
because when icsk_ack.quick=0 the !icsk_ack.pingpong value is
subsequently ignored as tcp_in_quickack_mode() checks both these
values. The condition for a quick ack to trigger requires
that both icsk_ack.quick != 0 and icsk_ack.pingpong=0. Currently
only icsk_ack.pingpong is controlled by the knob. But the
icsk_ack.quick value changes dynamically depending on heuristics.
The crux of the matter is that delayed acks still cannot be entirely
disabled even with the RTAX_QUICKACK per dst knob enabled. This
patch ensures that a quick ack is always sent when the RTAX_QUICKACK
per dst knob is turned on.

The "ip route change quickack 1" knob was recently added to enable
quickacks. It was modeled around the TCP_QUICKACK setsockopt() option.
This issue is that even with "ip route change quickack 1" enabled
we still see delayed ACKs under some conditions. It would be nice
to be able to completely disable delayed ACKs.

Here is an example:

# netstat -s|grep dela
    3 delayed acks sent

For all routes enable the knob

# ip route change quickack 1

Generate some traffic across a slow link and we still see the delayed
acks.

# netstat -s|grep dela
    106 delayed acks sent
    1 delayed acks further delayed because of locked socket

The issue is that both the "ip route change quickack 1" knob and
the TCP_QUICKACK option set the icsk_ack.pingpong variable to 0.
However at the business end in the __tcp_ack_snd_check() routine,
tcp_in_quickack_mode() checks that both icsk_ack.quick != 0
and icsk_ack.pingpong=0 in order to trigger a quickack. As
icsk_ack.quick is determined by heuristics it can be 0. When
that occurs the icsk_ack.pingpong value is ignored and a delayed
ACK is sent regardless.

This patch moves the RTAX_QUICKACK per dst check into the
tcp_in_quickack_mode() routine which ensures that a quickack is
always sent when the quickack knob is enabled for that dst.
Signed-off-by: NJon Maxwell <jmaxwell37@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Implement ndo_change_mtu: on MTU change, reallocate Rx ring bufs and signal
HW of new port MTU value.
Signed-off-by: NScott Feldman <sfeldma@gmail.com>
Reviewed-by: NSimon Horman <simon.horman@netronome.com>
Tested-by: NSimon Horman <simon.horman@netronome.com>
Acked-by: NJiri Pirko <jiri@resnulli.us>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use module_pci_driver for drivers whose init and exit functions
only register and unregister, respectively.

A simplified version of the Coccinelle semantic patch that performs
this transformation is as follows:

@a@
identifier f, x;
@@
-static f(...) { return pci_register_driver(&x); }

@b depends on a@
identifier e, a.x;
statement S;
@@
-static e(...) {
-pci_unregister_driver(&x);
-DBG_PRINT(INIT_DBG,"S");
- }

@c depends on a && b@
identifier a.f;
declarer name module_init;
@@
-module_init(f);

@d depends on a && b && c@
identifier b.e, a.x;
declarer name module_exit;
declarer name module_pci_driver;
@@
-module_exit(e);
+module_pci_driver(x);
Signed-off-by: NVaishali Thakkar <vthakkar1994@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If we don't have access to the new User GTS (T5+), use the old doorbell
mechanism; otherwise use the new BAR2 mechanism.
Signed-off-by: NHariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently "ALU_END_FROM_BE 32" and "ALU_END_FROM_LE 32" do not test if
the upper bits of the result are zeros (the arm64 JIT had such bugs).
Extend the two tests to catch this.
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@plumgrid.com>
Signed-off-by: NXi Wang <xi.wang@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Hariprasad Shenai says:

====================
Cleanup, T6 changes and register range update

This patch series adds the following:
Don't use entire L2T table, update register ranges for T6 adapter,
read stats for only available channels for T6 and enable cim_la dump for
T6 adapter also.

This patch series has been created against net-next tree and includes
patches on cxgb4 driver.

We have included all the maintainers of respective drivers. Kindly review
the change and let us know in case of any review comments.
====================
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NHariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Updating the driver to read the stats of only available channels. T6 and
later has only 2 channels
Signed-off-by: NHariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NHariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The driver was retrieving the parameters for the bounds of its
slice of the L2T from the firmware and then throwing those away and
using the entire table. This corrects that problem.
Signed-off-by: NHariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>