Jeff Kirsher says:

====================
Intel Wired LAN Driver Updates 2014-06-24

This series contains updates to i40e and i40evf.

Greg provides a patch to stop the VF device after setting its MAC address.
This is due to if the host VMM administrator has changed the VF device's MAC
address then the i40e driver needs to halt the VF device so that the
administrator will be forced to reload the VF driver so that the VF driver
will start using the newly assigned MAC address.

Shannon provides several patches for i40e, first makes prep_for_test() a
void function since the return value was being ignored for the most part
by all its callers.  Adds a log warning when the firmware's API minor/major
number is not what we expect to assist the user by informing them they
may need to update their NVM or SW.  Cleans up a stray print message
so that it is similar to other print messages.  Ensures to set the
WoL flag when setting LAA and allow the user to set LAA again.  So do
not short-circuit the LAA assignment when the driver thinks it has
already been done as it is possible that the user might want to force
the address setting again.  Provides a couple more LAA fixes to ensure
the LAA gets restored after resets.

Neerav provides a patch for i40e to add a PF reset when a malicious driver
event for the PF occurs.  As per the specification when the PF driver
receives a malicious driver event the queue that caused the event is
already stopped and it is expected that the function that owns the queue
will reset the queue, but in some cases it may not be possible to determine
the queue, so it is suggested to reset the whole function.

Carolyn fixes ethtool coalesce settings to allow 0 as a disable value and
adds message to user about invalid values.

Jesse removes a reserved type which was not removed from the code.

Catherine provides a patch to add the ability to enable/disable link from
set_link)restart_an() which will make it easy to toggle link without
calling set_phy_config() when no other link settings need to change.

Anjali provides a patch to ensure we do a PF reset on Tx hang and that
way we avoid any Malicious Driver Detect (MDD) events because of a Tx
queue disable failure.

v2:
 - fixed the un-needed return in patch 2 based on feedback from Sergei Shtylyov
 - added punctuation to print statements and code comment based on
   feedback from Sergei Shtylyov
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use bool instead of int as the return type.

All uses are tested with !.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

It seems overkill to use vmalloc() for typical listeners with less than
2048 hash buckets. Try kmalloc() and fallback to vmalloc() to reduce TLB
pressure.

Use kvfree() helper as it is now available.
Use ilog2() instead of a loop.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

It is trivial to add netpoll support to veth, since
it is not a stacked device, we don't need to setup and
clean up netpoll.
Reported-by: NStefan Priebe <s.priebe@profihost.ag>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Neil Horman <nhorman@tuxdriver.com>
Acked-by: NNeil Horman <nhorman@tuxdriver.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Acked-by: NNeil Horman <nhorman@tuxdriver.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Jitendra Kalsaria says:

====================
Rebranding changes for bnx2 and cnic driver

The following set of patches are for rebranding Broadcom bnx2 and cnic driver as QLogic drivers.
Please apply this series to net-next.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NJitendra Kalsaria <jitendra.kalsaria@qlogic.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

o QLogic has acquired the NetXtremeII products and drivers from Broadcom.
  This patch re-brands cnic driver as a QLogic driver
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

o QLogic has acquired the NetXtremeII products and drivers from Broadcom.
  This patch re-brands bnx2 driver as a QLogic driver
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fix checkpatch warning:
WARNING: kfree(NULL) is safe this check is probably not required

Cc: "David S. Miller" <davem@davemloft.net>
Cc: Grant Likely <grant.likely@linaro.org>
Cc: netdev@vger.kernel.org
Cc: Joe Perches <joe@perches.com>
Signed-off-by: NFabian Frederick <fabf@skynet.be>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Daniel Borkmann says:

====================
crc32 combine improvements

So almost a month passed, and I don't want this to get lost
somewhere. I have applied the feedback given at that time to
this set, rebased plus tested it against latest net-next. I
decided to route this via netdev as it improves performance
upon library code that provides library bits for SCTP, i.e.
for non-linear skb csum handling in IPVS. Thus, resending
this for George before it gets lost.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In case they help the compiler.
Signed-off-by: NGeorge Spelvin <linux@horizon.com>
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

So it gets discarded after the selftest.
Signed-off-by: NGeorge Spelvin <linux@horizon.com>
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There's no need for a full 32x32 matrix, when rows before the last are
just shifted copies of the rows after them.

There's still room for improvement (especially on X86 processors with
CRC32 and PCLMUL instructions), but this is a large step in the
right direction [which is in particular useful for its current user,
namely SCTP checksumming over multiple skb frags[] entries, i.e. in
IPVS balancing when other CRC32 offloads are not available].

The internal primitive is now called crc32_generic_shift and takes one
less argument; the XOR with crc2 is done in inline wrappers.
Signed-off-by: NGeorge Spelvin <linux@horizon.com>
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Bump versions.

Change-ID: I47fc3433240800cd823ff512f3015822277b0d20
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

When a Tx hang happens, usually the Tx queue disable fails. At
this point if we try to recover by a VSI reinit the HW gets
unhappy and we get a Malicious Driver Detect (MDD) event.
HW expects a PF reset if a queue disable fails, if we don't do a PF
reset and restart the queue we get an MDD.  This patch makes sure we
do a PF reset on Tx hang and that way we avoid any MDD because of Tx
queue disable failure.

Change-ID: I665ab6223577c788da857ee2132e733dc9a451e4
Signed-off-by: NAnjali Singhai Jain <anjali.singhai@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

The PF reset to clean up at the end of the remove is a nice thing
to do, but it also removes any LAA setting that Wake On LAN wants
for future wake up.

Change-ID: Ic090ec714df2d722281d11735cf75f2aa4432e2c
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

The LAA is lost on a reset, so be sure to replay it when rebuilding
the switch after any reset.

Change-ID: I6e643f9a59dfd899b6cbdf84d93b4bc9c37bb949
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Don't short-circuit the LAA assignment when the driver thinks it has
already been done - it is possible that the user might want to force
the address setting again.  At the same time, this requires a little
re-ordering of the filter management.

Change-ID: Ia0d71e3bc04edd7b68cf67edecc00abe7b9f6639
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Make sure the Firmware sets up the LAA as a Wake-On-LAN address.

Change-ID: I57b9acd8c288424fcfed0911053eb725c400b41c
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

The ability is already there in the fw and this will make it easy
to toggle link without calling set_phy_config when no other link
settings need to change.

Change-ID: I185567ae81776382ac145247e4eb1ee95f22382c
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

One of the PCTYPES that was moved to a reserved value
wasn't removed from the code.

Change-ID: I31fafe6d79c5f5128179979af5eaafa8c0cd62fe
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch fixes the i40e_set_coalesce function to allow 0 as a disable
value.  Also, added message to user about invalid value and provides valid
range.

Change-ID: I6c9ff11a9861f2045bd543745a3d132999ffbbd8
Signed-off-by: NCarolyn Wyborny <carolyn.wyborny@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This log print message will probably never be seen, but it needs to match
the "attempting to rebuild switch\n" log message a few lines above.

Change-ID: Ic3f5b4f67568d721cb02e826cf2cb33847f51c11
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

If the firmware's API minor number is larger than the one expected, log
a warning and recommend driver SW update.
If the firmware's API major or minor number is smaller then the one expected
(n for major, n or n-1 for minor), log a warning and recommend NVM update.

Change-ID: If0b887e055478f8e435ba7fa28113b63a6f1bb35
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

As per the spec when the PF driver receives a Malicious driver event
the queue that caused the event is already stopped and it is expected
that the function that owns the queue will reset the queue.
In some cases it may not be possible to determine the queue and it is
suggested to reset the whole function.

This patch takes the later approach when the event is owned by the PF
that owns it.

Change-ID: I40f9764a6a5e068c0ef8438db00c5aa9c2c6c1c8
Signed-off-by: NNeerav Parikh <neerav.parikh@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

The return from i40e_prep_for_reset() was being ignored by almost all
its callers.  The one place it wasn't ignored could have caused a silent
and confusing failure of the driver to finish a reset.  Since we really
are doing a rebuild anyway, ignore this last case as well and simply
make the function a void type.

Change-ID: Ia4fed7f903d39a6c47c5722625a53e59c3f7ed53
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

If the host VMM administrator has changed the VF device's MAC address then
the i40e driver needs to halt the VF device so that the administrator will
be forced to reload the VF driver.  This will cause the VF driver to start
using the newly assigned MAC address.  This brings the i40e driver into
conformance with operational characteristics of other Intel SR-IOV
featured drivers.

Change-ID: Ic7242cceb8287dd2cb72fb1f3166a032a28bf88a
Signed-off-by: NGreg Rose <gregory.v.rose@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Hi,
 This patch fixes warnings generated by sparse as pointed out by kbuild test
 robot, please apply to net-next. Applies on top of
commit 79631c89 ("trivial: net/irda/irlmp.c:
Fix closing brace followed by if")
-Anish

v2: cleanup submission as per davem's feedback

Fixes: 76bcb31e ("cxgb4 : Add DCBx support codebase  and dcbnl_ops")
Signed-off-by: NAnish Bhatt <anish@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NRasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Govindarajulu Varadarajan says:

====================
enic updates

This series fixes minor bugs and adds new features like Accelerated RFS,
busy_poll, tx clean-up in napi_poll.

v3:
* While doing tx cleanup in napi, ignore budget and clean up all desc possible.

v2:
* Fix #ifdef coding style issue in '[PATCH 4/8] enic: alloc/free rx_cpu_rmap'
  And [PATCH 5/8] enic: Add Accelerated RFS support'
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Till now enic had been doing tx clean in isr.

Using napi infrastructure to move the tx clean up out of isr to softirq.
Now, wq isr schedules napi poll. In enic_poll_msix_wq we clean up the tx queus.

This is applicable only on MSIX. In INTx and MSI we use single napi to clean
both rx & tx queues.
Signed-off-by: NGovindarajulu Varadarajan <_govind@gmx.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds support for low latency busy_poll.

* Introduce drivers ndo_busy_poll function enic_busy_poll, which is called by
socket waiting for data.

* Introduce locking between napi_poll nad busy_poll

* enic_busy_poll cleans up all the rx pkts possible. While in busy_poll, rq
holds the state ENIC_POLL_STATE_POLL. While in napi_poll, rq holds the state
ENIC_POLL_STATE_NAPI.

* in napi_poll we return if we are in busy_poll. Incase of INTx & msix, we just
service wq and return if busy_poll is going on.
Signed-off-by: NGovindarajulu Varadarajan <_govind@gmx.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We were experiencing occasional "BUG: scheduling while atomic" splats
in our testing. Enabling DEBUG_SPINLOCK and DEBUG_LOCKDEP in the kernel
exposed a lockdep in the enic driver.

enic 0000:0b:00.0 eth2: Link UP

======================================================
[ INFO: SOFTIRQ-safe -> SOFTIRQ-unsafe lock order detected ]
3.12.0-rc1.x86_64-dbg+ #2 Tainted: GF       W
------------------------------------------------------
NetworkManager/4209 [HC0[0]:SC0[2]:HE1:SE0] is trying to acquire:
(&(&enic->devcmd_lock)->rlock){+.+...}, at: [<ffffffffa026b7e4>] enic_dev_packet_filter+0x44/0x90 [enic]

The fix was to replace spin_lock with spin_lock_bh for the enic
devcmd_lock, so that soft irqs would be disabled while the lock
is held.
Signed-off-by: NSujith Sankar <ssujith@cisco.com>
Signed-off-by: NTony Camuso <tcamuso@redhat.com>
Signed-off-by: NGovindarajulu Varadarajan <_govind@gmx.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds supports for Accelerated Receive Flow Steering.

When the desired rx is different from current rq, for a flow, kernel calls the
driver function enic_rx_flow_steer(). enic_rx_flow_steer adds a IP-TCP/UDP
hardware filter.

Driver registers a timer function enic_flow_may_expire. This function is called
every HZ/4 seconds. In this function we check if the added filter has expired
by calling rps_may_expire_flow(). If the flow has expired, it removes the hw
filter.

As of now adaptor supports only IPv4 - TCP/UDP filters.
Signed-off-by: NGovindarajulu Varadarajan <_govind@gmx.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

rx_cpu_rmap provides the reverse irq cpu affinity. This patch allocates and
sets drivers netdev->rx_cpu_rmap accordingly.

rx_cpu_rmap is set in enic_request_intr() which is called by enic_open and
rx_cpu_rmap is freed in enic_free_intr() which is called by enic_stop.

This is used by Accelerated RFS.
Signed-off-by: NGovindarajulu Varadarajan <_govind@gmx.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds interface to add and delete IP 5 tuple filter. This interface
is used by Accelerated RFS code to steer a flow to corresponding receive
queue.

As of now adaptor supports only ipv4 + tcp/udp packet steering.
Signed-off-by: NGovindarajulu Varadarajan <_govind@gmx.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Hardware (in readq(&devcmd->args[0])) returns positive number in case of error.
But _vnic_dev_cmd should return a negative value in case of error.
Signed-off-by: NGovindarajulu Varadarajan <_govind@gmx.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

skb_flow_dissect() dissects only transport header type in ip_proto. It dose not
give any information about IPv4 or IPv6.

This patch adds new member, n_proto, to struct flow_keys. Which records the
IP layer type. i.e IPv4 or IPv6.

This can be used in netdev->ndo_rx_flow_steer driver function to dissect flow.

Adding new member to flow_keys increases the struct size by around 4 bytes.
This causes BUILD_BUG_ON(sizeof(qcb->data) < sz); to fail in
qdisc_cb_private_validate()

So increase data size by 4
Signed-off-by: NGovindarajulu Varadarajan <_govind@gmx.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Varka Bhadram says:

====================
Driver for TI CC2520 Radio

changes since v6:
	- proper releasing of resources
	- changed pr_debug() to dev_dbg()
	- removed unwanted header file
	- changed is_tx type to bool

changes since v5:
	- added saddr and panc_changed functionality
	- rework on lqi field
	- improvements in checkings for reception
	- modification in debug messages

changes since v4:
	- feature to write into cc2520 RAM
	- added h/w address filtering

changes since v3:
	- drop the unused varible

changes since v2:
	- drop spi mode dt binding

changes since v1:
	- improvement in gpio setup
	- changed len pointer to len variable

changes for v1:
	- improvements in the locking mechanism in Tx and SFD ISR
	- proper checkings for GPIO pins
	- avoids the memory leak for priv
	- used devm_* API's
	- moved the code from header file to .c file
	- removed cc2520_unregister()
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>