Hannes Frederic Sowa says:

====================
This series implements support for delaying the initialization of secret
keys, e.g. used for hashing, for as long as possible. This functionality
is implemented by a new macro, net_get_random_bytes.

I already used it to protect the socket hashes, the syncookie secret
(most important) and the tcp_fastopen secrets.

Changelog:
v2) Use static_keys in net_get_random_once to have as minimal impact to
    the fast-path as possible.
v3) added patch "static_key: WARN on usage before jump_label_init was called":
    Patch "x86/jump_label: expect default_nop if static_key gets enabled
    on boot-up" relaxes the checks for using static_key primitives before
    jump_label_init. So tighten them first.
v4) Update changelog on the patch "static_key: WARN on usage before
    jump_label_init was called"

Included patches:
 ipv4: split inet_ehashfn to hash functions per compilation unit
 ipv6: split inet6_ehashfn to hash functions per compilation unit
 static_key: WARN on usage before jump_label_init was called
 x86/jump_label: expect default_nop if static_key gets enabled on boot-up
 net: introduce new macro net_get_random_once
 inet: split syncookie keys for ipv4 and ipv6 and initialize with net_get_random_once
 inet: convert inet_ehash_secret and ipv6_hash_secret to net_get_random_once
 tcp: switch tcp_fastopen key generation to net_get_random_once
 net: switch net_secret key generation to net_get_random_once
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Cc: Eric Dumazet <edumazet@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Changed key initialization of tcp_fastopen cookies to net_get_random_once.

If the user sets a custom key net_get_random_once must be called at
least once to ensure we don't overwrite the user provided key when the
first cookie is generated later on.

Cc: Yuchung Cheng <ycheng@google.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Initialize the ehash and ipv6_hash_secrets with net_get_random_once.

Each compilation unit gets its own secret now:
  ipv4/inet_hashtables.o
  ipv4/udp.o
  ipv6/inet6_hashtables.o
  ipv6/udp.o
  rds/connection.o

The functions still get inlined into the hashing functions. In the fast
path we have at most two (needed in ipv6) if (unlikely(...)).

Cc: Eric Dumazet <edumazet@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

This patch splits the secret key for syncookies for ipv4 and ipv6 and
initializes them with net_get_random_once. This change was the reason I
did this series. I think the initialization of the syncookie_secret is
way to early.

Cc: Florian Westphal <fw@strlen.de>
Cc: Eric Dumazet <edumazet@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

net_get_random_once is a new macro which handles the initialization
of secret keys. It is possible to call it in the fast path. Only the
initialization depends on the spinlock and is rather slow. Otherwise
it should get used just before the key is used to delay the entropy
extration as late as possible to get better randomness. It returns true
if the key got initialized.

The usage of static_keys for net_get_random_once is a bit uncommon so
it needs some further explanation why this actually works:

=== In the simple non-HAVE_JUMP_LABEL case we actually have ===
no constrains to use static_key_(true|false) on keys initialized with
STATIC_KEY_INIT_(FALSE|TRUE). So this path just expands in favor of
the likely case that the initialization is already done. The key is
initialized like this:

___done_key = { .enabled = ATOMIC_INIT(0) }

The check

                if (!static_key_true(&___done_key))                     \

expands into (pseudo code)

                if (!likely(___done_key > 0))

, so we take the fast path as soon as ___done_key is increased from the
helper function.

=== If HAVE_JUMP_LABELs are available this depends ===
on patching of jumps into the prepared NOPs, which is done in
jump_label_init at boot-up time (from start_kernel). It is forbidden
and dangerous to use net_get_random_once in functions which are called
before that!

At compilation time NOPs are generated at the call sites of
net_get_random_once. E.g. net/ipv6/inet6_hashtable.c:inet6_ehashfn (we
need to call net_get_random_once two times in inet6_ehashfn, so two NOPs):

      71:       0f 1f 44 00 00          nopl   0x0(%rax,%rax,1)
      76:       0f 1f 44 00 00          nopl   0x0(%rax,%rax,1)

Both will be patched to the actual jumps to the end of the function to
call __net_get_random_once at boot time as explained above.

arch_static_branch is optimized and inlined for false as return value and
actually also returns false in case the NOP is placed in the instruction
stream. So in the fast case we get a "return false". But because we
initialize ___done_key with (enabled != (entries & 1)) this call-site
will get patched up at boot thus returning true. The final check looks
like this:

                if (!static_key_true(&___done_key))                     \
                        ___ret = __net_get_random_once(buf,             \

expands to

                if (!!static_key_false(&___done_key))                     \
                        ___ret = __net_get_random_once(buf,             \

So we get true at boot time and as soon as static_key_slow_inc is called
on the key it will invert the logic and return false for the fast path.
static_key_slow_inc will change the branch because it got initialized
with .enabled == 0. After static_key_slow_inc is called on the key the
branch is replaced with a nop again.

=== Misc: ===
The helper defers the increment into a workqueue so we don't
have problems calling this code from atomic sections. A seperate boolean
(___done) guards the case where we enter net_get_random_once again before
the increment happend.

Cc: Ingo Molnar <mingo@redhat.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Eric Dumazet <edumazet@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

net_get_random_once(intrduced in the next patch) uses static_keys in
a way that they get enabled on boot-up instead of replaced with an
ideal_nop. So check for default_nop on initial enabling.

Other architectures don't check for this.

Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Eric Dumazet <edumazet@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: x86@kernel.org
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Usage of the static key primitives to toggle a branch must not be used
before jump_label_init() is called from init/main.c. jump_label_init
reorganizes and wires up the jump_entries so usage before that could
have unforeseen consequences.

Following primitives are now checked for correct use:
* static_key_slow_inc
* static_key_slow_dec
* static_key_slow_dec_deferred
* jump_label_rate_limit

The x86 architecture already checks this by testing if the default_nop
was already replaced with an optimal nop or with a branch instruction. It
will panic then. Other architectures don't check for this.

Because we need to relax this check for the x86 arch to allow code to
transition from default_nop to the enabled state and other architectures
did not check for this at all this patch introduces checking on the
static_key primitives in a non-arch dependent manner.

All checked functions are considered slow-path so the additional check
does no harm to performance.

The warnings are best observed with earlyprintk.

Based on a patch from Andi Kleen.

Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

This patch splits the inet6_ehashfn into separate ones in
ipv6/inet6_hashtables.o and ipv6/udp.o to ease the introduction of
seperate secrets keys later.

Cc: Eric Dumazet <edumazet@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

This duplicates a bit of code but let's us easily introduce
separate secret keys later. The separate compilation units are
ipv4/inet_hashtabbles.o, ipv4/udp.o and rds/connection.o.

Cc: Eric Dumazet <edumazet@google.com>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Eric Dumazet says:

====================
net: Implement GSO/TSO support for IPIP

This patch serie implements GSO/TSO support for IPIP

David, please note it applies after "ipv4: gso: send_check() & segment() cleanups"
( http://patchwork.ozlabs.org/patch/284714/ )

Broadcom bnx2x driver is now enabled for TSO support of IPIP traffic

Before patch :

lpq83:~# ./netperf -H 7.7.9.84 -Cc
MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 7.7.9.84 () port 0 AF_INET
Recv   Send    Send                          Utilization       Service Demand
Socket Socket  Message  Elapsed              Send     Recv     Send    Recv
Size   Size    Size     Time     Throughput  local    remote   local   remote
bytes  bytes   bytes    secs.    10^6bits/s  % S      % S      us/KB   us/KB

 87380  16384  16384    10.00      3357.88   5.09     3.70     2.983   2.167

After patch :

lpq83:~# ./netperf -H 7.7.9.84 -Cc
MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 7.7.9.84 () port 0 AF_INET
Recv   Send    Send                          Utilization       Service Demand
Socket Socket  Message  Elapsed              Send     Recv     Send    Recv
Size   Size    Size     Time     Throughput  local    remote   local   remote
bytes  bytes   bytes    secs.    10^6bits/s  % S      % S      us/KB   us/KB

 87380  16384  16384    10.00      8532.40   2.55     7.73     0.588   1.781
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

bnx2x driver already handles TSO for GRE, current code
is the same for IPIP.

Performance results : (Note we are now limited by receiver,
as it does not support GRO for IPIP yet)

Before patch :

lpq83:~# ./netperf -H 7.7.9.84 -Cc
MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 7.7.9.84 () port 0 AF_INET
Recv   Send    Send                          Utilization       Service Demand
Socket Socket  Message  Elapsed              Send     Recv     Send    Recv
Size   Size    Size     Time     Throughput  local    remote   local   remote
bytes  bytes   bytes    secs.    10^6bits/s  % S      % S      us/KB   us/KB

 87380  16384  16384    10.00      7710.19   4.52     6.62     1.152   1.687

After patch :

lpq83:~# ./netperf -H 7.7.9.84 -Cc
MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 7.7.9.84 () port 0 AF_INET
Recv   Send    Send                          Utilization       Service Demand
Socket Socket  Message  Elapsed              Send     Recv     Send    Recv
Size   Size    Size     Time     Throughput  local    remote   local   remote
bytes  bytes   bytes    secs.    10^6bits/s  % S      % S      us/KB   us/KB

 87380  16384  16384    10.00      8532.40   2.55     7.73     0.588   1.781
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Now inet_gso_segment() is stackable, its relatively easy to
implement GSO/TSO support for IPIP

Performance results, when segmentation is done after tunnel
device (as no NIC is yet enabled for TSO IPIP support) :

Before patch :

lpq83:~# ./netperf -H 7.7.9.84 -Cc
MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 7.7.9.84 () port 0 AF_INET
Recv   Send    Send                          Utilization       Service Demand
Socket Socket  Message  Elapsed              Send     Recv     Send    Recv
Size   Size    Size     Time     Throughput  local    remote   local   remote
bytes  bytes   bytes    secs.    10^6bits/s  % S      % S      us/KB   us/KB

 87380  16384  16384    10.00      3357.88   5.09     3.70     2.983   2.167

After patch :

lpq83:~# ./netperf -H 7.7.9.84 -Cc
MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 7.7.9.84 () port 0 AF_INET
Recv   Send    Send                          Utilization       Service Demand
Socket Socket  Message  Elapsed              Send     Recv     Send    Recv
Size   Size    Size     Time     Throughput  local    remote   local   remote
bytes  bytes   bytes    secs.    10^6bits/s  % S      % S      us/KB   us/KB

 87380  16384  16384    10.00      7710.19   4.52     6.62     1.152   1.687
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

In order to support GSO on IPIP, we need to make
inet_gso_segment() stackable.

It should not assume network header starts right after mac
header.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This patch makes gre_handle_offloads() more generic
and rename it to iptunnel_handle_offloads()

This will be used to add GSO/TSO support to IPIP tunnels.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

While implementing GSO/TSO support for IPIP, I found skb_segment()
was assuming network header was immediately following mac header.

Its not really true in the case inet_gso_segment() is stacked :
By the time tcp_gso_segment() is called, network header points
to the inner IP header.

Let's instead assume nothing and pick the current offsets found in
original skb, we have skb_headers_offset_update() helper for that.

Also move the csum_start update inside skb_headers_offset_update()
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

ipv6_gso_send_check() and ipv6_gso_segment() are called by
skb_mac_gso_segment() under rcu lock, no need to use
rcu_read_lock() / rcu_read_unlock()
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Currently we log a message whenever pcie_enable_error_reporting fails.
The message clutters up logs, especially when we don't support it for VFs.
Instead enable this only for PFs and log a message when the call succeeds.
Signed-off-by: Ajit Khaparde <ajit.khaparde@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

There are a mix of function prototypes with and without extern
in the kernel sources.  Standardize on not using extern for
function prototypes.

Function prototypes don't need to be written with extern.
extern is assumed by the compiler.  Its use is as unnecessary as
using auto to declare automatic/local variables in a block.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

There are a mix of function prototypes with and without extern
in the kernel sources.  Standardize on not using extern for
function prototypes.

Function prototypes don't need to be written with extern.
extern is assumed by the compiler.  Its use is as unnecessary as
using auto to declare automatic/local variables in a block.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

There are a mix of function prototypes with and without extern
in the kernel sources.  Standardize on not using extern for
function prototypes.

Function prototypes don't need to be written with extern.
extern is assumed by the compiler.  Its use is as unnecessary as
using auto to declare automatic/local variables in a block.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

There are a mix of function prototypes with and without extern
in the kernel sources.  Standardize on not using extern for
function prototypes.

Function prototypes don't need to be written with extern.
extern is assumed by the compiler.  Its use is as unnecessary as
using auto to declare automatic/local variables in a block.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

inet_gso_segment() and inet_gso_send_check() are called by
skb_mac_gso_segment() under rcu lock, no need to use
rcu_read_lock() / rcu_read_unlock()

Avoid calling ip_hdr() twice per function.

We can use ip_send_check() helper.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

It can be called from the module init function, so it cannot
be in the exit section.
Signed-off-by: David S. Miller <davem@davemloft.net>

Jiri Pirko says:

====================
bonding: introduce bonding options Netlink support

This patchset basically allows "mode" and "active_slave" bonding options
to be propagated and set up via standart RT Netlink interface.

In future other options can be easily added as well.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>

no longer needed since bond_option_active_slave_set() can be used
instead.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>

Do a bit of refactoring on the way.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>

Randy found that if network namespace not enabled then
nd_net does not exist and would cause compilation failure.

This is handled correctly by using the dev_net() macro.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Remove the specialized code in __tcp_retransmit_skb() that tries to trim
any ACKed payload preceding a FIN before we retransmit (this was added
in 1999 in v2.2.3pre3). This trimming code was made unreachable by the
more general code added above it that uses tcp_trim_head() to trim any
ACKed payload, with or without a FIN (this was added in "[NET]: Add
segmentation offload support to TCP." in 2002 circa v2.5.33).
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Acked-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Under certain low traffic conditions, the single core
devices with multiple Rx/Tx queues (MQ mode) may reach
soft lockup due to gfar_poll not returning in proper time.
The following exception was obtained using iperf on a 100Mbit
half-duplex link, for a p1010 single core device:

BUG: soft lockup - CPU#0 stuck for 23s! [iperf:2847]
Modules linked in:
CPU: 0 PID: 2847 Comm: iperf Not tainted 3.12.0-rc3 #16
task: e8bf8000 ti: eeb16000 task.ti: ee646000
NIP: c0255b6c LR: c0367ae8 CTR: c0461c18
REGS: eeb17e70 TRAP: 0901   Not tainted  (3.12.0-rc3)
MSR: 00029000 <CE,EE,ME>  CR: 44228428  XER: 20000000

GPR00: c0367ad4 eeb17f20 e8bf8000 ee01f4b4 00000008 ffffffff ffffffff
00000000
GPR08: 000000c0 00000008 000000ff ffffffc0 000193fe
NIP [c0255b6c] find_next_bit+0xb8/0xc4
LR [c0367ae8] gfar_poll+0xc8/0x1d8
Call Trace:
[eeb17f20] [c0367ad4] gfar_poll+0xb4/0x1d8 (unreliable)
[eeb17f70] [c0422100] net_rx_action+0xa4/0x158
[eeb17fa0] [c003ec6c] __do_softirq+0xcc/0x17c
[eeb17ff0] [c000c28c] call_do_softirq+0x24/0x3c
[ee647cc0] [c0004660] do_softirq+0x6c/0x94
[ee647ce0] [c003eb9c] local_bh_enable+0x9c/0xa0
[ee647cf0] [c0454fe8] tcp_prequeue_process+0xa4/0xdc
[ee647d10] [c0457e44] tcp_recvmsg+0x498/0x96c
[ee647d80] [c047b630] inet_recvmsg+0x40/0x64
[ee647da0] [c040ca8c] sock_recvmsg+0x90/0xc0
[ee647e30] [c040edb8] SyS_recvfrom+0x98/0xfc

To prevent this, the outer while() loop has been removed
allowing gfar_poll() to return faster even if there's
still budget left.  Also, there's no need to recompute
the budget per Rx queue anymore.
Signed-off-by: Claudiu Manoil <claudiu.manoil@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The rtmsg_fib function doesn't modify this argument so mark
it const.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The current test works fine in practice.  The "amount" variable is
actually used as a boolean so negative values or any non-zero values
count as "true".  However since we don't allow numbers greater than one,
let's not allow negative numbers either.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

fib_table_lookup has included the rcu lock protection.
Signed-off-by: baker.zhang <baker.kernel@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Steffen Klassert says:

====================
1) Don't use a wildcard SA if a more precise one is in acquire state,
   from Fan Du.

2) Simplify the SA lookup when using wildcard source. We need to check
   only the destination in this case, from Fan Du.

3) Add a receive path hook for IPsec virtual tunnel interfaces
   to xfrm6_mode_tunnel.

4) Add support for IPsec virtual tunnel interfaces to ipv6.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Himanshu Madhani says:

====================
qlcnic: ethtool enhancements and code cleanup.

This patch series contains

o updates to ethtool for pause settings and enhance
  register dump to display mask and ring indices.
o cleanup in DCB code and remove redundant eSwitch enablement command.
o fixed firmware dump collection logic to skip unknown entries.

Changes from v3 -> v4
o Dropped patch for Tx queue validation to be submitted in net.

Changes from v2 -> v3

o Updated patch to print informational messages as per Joe Perches's comment.

Changes from v1 -> v2

o Dropped patch to register device if adapter is in FAILED state for more rework.
o Updated patch to display ring indices via ethtool per Ben Hutchings's comment.
o Update patch for DCB cleanup per Stephen Hemminger's comment.

Please apply to net-next.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>