Currently there is no way of querying whether a filter is
offloaded to HW or not when using "both" policy (where none
of skip_sw or skip_hw flags are set by user-space).

Add two new flags, "in hw" and "not in hw" such that user
space can determine if a filter is actually offloaded to
hw or not. The "in hw" UAPI semantics was chosen so it's
similar to the "skip hw" flag logic.

If none of these two flags are set, this signals running
over older kernel.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Amir Vadai <amir@vadai.me>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The classifier flags are not dumped to user-space, do that.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Yotam Gigi <yotamg@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Dump the classifier flags only if non zero and make sure to check
the return status of the handler that puts them into the netlink msg.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The ethtool api {get|set}_settings is deprecated.
We move this driver to new api {get|set}_link_ksettings.

As I don't have the hardware, I'd be very pleased if
someone may test this patch.
Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The ale is a property of cpsw, so change dev to cpsw->dev,
aka pdev->dev, to be consistent.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

The ethtool api {get|set}_settings is deprecated.
We move this driver to new api {get|set}_link_ksettings.

As I don't have the hardware, I'd be very pleased if
someone may test this patch.
Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Dmitry Torokhov says:

====================
PTP attribute handling cleanup

PTP core was creating some attributes, such as "period" and "fifo", and the
entire "pins" attribute group, after creating class deevice, which creates
a race for userspace: uevent may arrive before all attributes are created.

This series of patches switches PTP to use is_visible() to control
visibility of attributes in a group, and device_create_with_groups() to
ensure that attributes are created before we notify userspace of a new
device.

v2:
- added Richard's acked-by to patch #1
- removed use of kmalloc_array in favor of kcalloc in patch #2 at
  Richard's request
- added a cover letter

v1:
- initial patch set
====================
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Let's switch to using device_create_with_groups(), which will allow us to
create "pins" attribute group together with the rest of ptp device
attributes, and before userspace gets notified about ptp device creation.
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Instead of creating selected attributes after the device is created (and
after userspace potentially seen uevent), lets use attribute group
is_visible() method to control which attributes are shown. This will allow
us to create all attributes (except "pins" group, which will be taken care
of later) before userspace gets notified about new ptp class device.
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

kcalloc is more semantically correct when allocating arrays of objects, and
overflow-safe.
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

We do not need explicitly call dev_set_drvdata(), as it is done for us by
device_create().
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

All rx and rx netdev interrupts are handled by respectively
by mlx4_en_rx_irq() and mlx4_en_tx_irq() which simply schedule a NAPI.

But mlx4_eq_int() also fires a tasklet to service all items that were
queued via mlx4_add_cq_to_tasklet(), but this handler was not called
unless user cqe was handled.

This is very confusing, as "mpstat -I SCPU ..." show huge number of
tasklet invocations.

This patch saves this overhead, by carefully firing the tasklet directly
from mlx4_add_cq_to_tasklet(), removing four atomic operations per IRQ.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Tariq Toukan <tariqt@mellanox.com>
Cc: Saeed Mahameed <saeedm@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Steffen Klassert says:

====================
pull request (net-next): ipsec-next 2017-02-16

1) Make struct xfrm_input_afinfo const, nothing writes to it.
   From Florian Westphal.

2) Remove all places that write to the afinfo policy backend
   and make the struct const then.
   From Florian Westphal.

3) Prepare for packet consuming gro callbacks and add
   ESP GRO handlers. ESP packets can be decapsulated
   at the GRO layer then. It saves a round through
   the stack for each ESP packet.

Please note that this has a merge coflict between commit

63fca65d ("net: add confirm_neigh method to dst_ops")

from net-next and

3d7d25a6 ("xfrm: policy: remove garbage_collect callback")
a2817d8b ("xfrm: policy: remove family field")

from ipsec-next.

The conflict can be solved as it is done in linux-next.

Please pull or let me know if there are problems.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Jeff Kirsher says:

====================
10GbE Intel Wired LAN Driver Updates 2017-02-16

This series contains updates to ixgbe only.

Tony updates the driver to advertise 2.5Gb and 5.0Gb if the adapter
supports it.

Stephen Hemminger renames our dcbnl_ops since it is global to
ixgbe_dcbnl_ops to avoid namespace issues.

Mark updates the driver version based on the recent changes.

Alex has the remainder of the changes, starting with consolidating
functions that represent logical steps in the receive process so we can
later update them more easily (and align with igb).  Modify the receive
path to only synchronize the length of the frame versus the entire buffer.
Provided performance improvements by adding support for
DMA_ATTR_SKIP_CPU_SYNC and DMA_ATTR_WEAK_ORDERING.  Also made additional
performance gains by batching the page count updates instead of doing
them one at a time.  Adjusted the receive path to use 3k buffers with
8k backing them in order to support build_skb with jumbo frames.  Made
additional driver improvements by using the length of the packet instead
of the DD status to determine if a new descriptor is ready to be
processed, which cuts down on reads.  To reduce code duplication, pulled
apart the receive path into separate functions.  Added support for
providing a buffer with headroom and tailroom to allow for shared info
for NET_SKB_PAD and NET_IP_ALIGN.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Remove variable rxq_info and also remove redundant assignment
to it.
Signed-off-by: Ganesh Goudar <ganeshgr@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Ganesh Goudar <ganeshgr@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Make max number of supported tc u32 links equal to max number of filters
supported by hardware.
Signed-off-by: Arjun V <arjun@chelsio.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Signed-off-by: Casey Leedom <leedom@chelsio.com>
Signed-off-by: Ganesh Goudar <ganeshgr@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Pull media fix from Mauro Carvalho Chehab:
 "A regression fix that makes the Siano driver to work again after the
  CONFIG_VMAP_STACK change"

* tag 'media/v4.10-5' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media:
  [media] siano: make it work again with CONFIG_VMAP_STACK

Flags (PIPE_BUF_FLAG_PACKET, PIPE_BUF_FLAG_GIFT) could remain on the
unused part of the pipe ring buffer.  Previously splice_to_pipe() left
the flags value alone, which could result in incorrect behavior.

Uninitialized flags appears to have been there from the introduction of
the splice syscall.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Cc: <stable@vger.kernel.org> # 2.6.17+
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Pull fuse fixes from Miklos Szeredi:
 "Fix a use after free bug introduced in 4.2 and using an uninitialized
  value introduced in 4.9"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
  fuse: fix uninitialized flags in pipe_buffer
  fuse: fix use after free issue in fuse_dev_do_read()

Pull PCI fix from Bjorn Helgaas:
 "Add back pcie_pme_remove() so we free the IRQ when removing PCIe port
  devices; previously the leaked IRQ caused an MSI BUG_ON"

* tag 'pci-v4.10-fixes-4' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
  PCI/PME: Restore pcie_pme_driver.remove

Pull networking fixes from David Miller:

 1) In order to avoid problems in the future, make cgroup bpf overriding
    explicit using BPF_F_ALLOW_OVERRIDE. From Alexei Staovoitov.

 2) LLC sets skb->sk without proper skb->destructor and this explodes,
    fix from Eric Dumazet.

 3) Make sure when we have an ipv4 mapped source address, the
    destination is either also an ipv4 mapped address or
    ipv6_addr_any(). Fix from Jonathan T. Leighton.

 4) Avoid packet loss in fec driver by programming the multicast filter
    more intelligently. From Rui Sousa.

 5) Handle multiple threads invoking fanout_add(), fix from Eric
    Dumazet.

 6) Since we can invoke the TCP input path in process context, without
    BH being disabled, we have to accomodate that in the locking of the
    TCP probe. Also from Eric Dumazet.

 7) Fix erroneous emission of NETEVENT_DELAY_PROBE_TIME_UPDATE when we
    aren't even updating that sysctl value. From Marcus Huewe.

 8) Fix endian bugs in ibmvnic driver, from Thomas Falcon.

[ This is the second version of the pull that reverts the nested
  rhashtable changes that looked a bit too scary for this late in the
  release  - Linus ]

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (27 commits)
  rhashtable: Revert nested table changes.
  ibmvnic: Fix endian errors in error reporting output
  ibmvnic: Fix endian error when requesting device capabilities
  net: neigh: Fix netevent NETEVENT_DELAY_PROBE_TIME_UPDATE notification
  net: xilinx_emaclite: fix freezes due to unordered I/O
  net: xilinx_emaclite: fix receive buffer overflow
  bpf: kernel header files need to be copied into the tools directory
  tcp: tcp_probe: use spin_lock_bh()
  uapi: fix linux/if_pppol2tp.h userspace compilation errors
  packet: fix races in fanout_add()
  ibmvnic: Fix initial MTU settings
  net: ethernet: ti: cpsw: fix cpsw assignment in resume
  kcm: fix a null pointer dereference in kcm_sendmsg()
  net: fec: fix multicast filtering hardware setup
  ipv6: Handle IPv4-mapped src to in6addr_any dst.
  ipv6: Inhibit IPv4-mapped src address on the wire.
  net/mlx5e: Disable preemption when doing TC statistics upcall
  rhashtable: Add nested tables
  tipc: Fix tipc_sk_reinit race conditions
  gfs2: Use rhashtable walk interface in glock_hash_walk
  ...

Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Fixes: d82718e3 ("fuse_dev_splice_read(): switch to add_to_pipe()")
Cc: <stable@vger.kernel.org> # 4.9+

This patch makes it so that we don't need to bother with clearing the
memory out for the descriptor rings.  The general idea is to only free
buffers associated with buffers in use which are located between the
next_to_clean and next_to_use or next_to_alloc values.  Everything outside
of those regions can be safely ignored since they should have no buffers
associated with them.

The advantage to doing things this way is that is should speed up bring-up
and tear-down of the rings.  Specifically we can avoid the 512 or more
cycles required to memset the rings in tear-down.  In the bring-up phase we
then clear the memory as a part of initialization.  The general idea is
that the clearing in initialization can act as a prefetch of sorts for the
buffer info structures so they are in the local CPU when we go to populate
them.  This should help to improve overall time needed to perform a
suspend/resume.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch adds build_skb support to the Rx path.  There are several
advantages to this change.

1.  It avoids the memcpy and skb->head allocation for small packets which
    improves performance by about 5% in my tests.
2.  It avoids the memcpy, skb->head allocation, and eth_get_headlen
    for larger packets improving performance by about 10% in my tests.
3.  For VXLAN packets it allows the full header to be in skb->data which
    improves the performance by as much as 30% in some of my tests.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Since there are potential drawbacks to the new Rx allocation approach I
thought it best to add a "chicken bit" so that we can turn the feature off
if in the event that a problem is found.

It also provides a means of validating the legacy Rx path in the event that
we are forced to fall back.  At some point in the future when we are
convinced we don't need it anymore we might be able to drop the legacy-rx
flag.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch adds support for providing a buffer with headroom and tailroom
to allow for shared info, NET_SKB_PAD, and NET_IP_ALIGN.  With this
combined with the DMA changes we can start using build_skb to build frames
around an incoming Rx buffer instead of having to memcpy the headers.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

We are going to be expanding the number of Rx paths in the driver.  Instead
of duplicating all that code I am pulling it apart into separate functions
so that we don't have so much code duplication.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

This change makes it so that we use the length of the packet instead of the
DD status bit to determine if a new descriptor is ready to be processed.
The obvious advantage is that it cuts down on reads as we don't really even
need the DD bit if going from a 0 to a non-zero value on size is enough to
inform us that the packet has been completed.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

In order to support build_skb with jumbo frames it will be necessary to use
3K buffers for the Rx path with 8K pages backing them.  This is needed on
architectures that implement 4K pages because we can't support 2K buffers
plus padding in a 4K page.

In the case of systems that support page sizes larger than 4K the 3K
attribute will only be applied to FCoE as we can fall back to using just 2K
buffers and adding the padding.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Batch the page count updates instead of doing them one at a time.  By doing
this we can improve the overall performance as the atomic increment
operations can be expensive due to the fact that on x86 they are locked
operations which can cause stalls.  By doing bulk updates we can
consolidate the stall which should help to improve the overall receive
performance.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Acked-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch adds support for DMA_ATTR_SKIP_CPU_SYNC and
DMA_ATTR_WEAK_ORDERING.  By enabling both of these for the Rx path we are
able to see performance improvements on architectures that implement either
one due to the fact that page mapping and unmapping only has to sync what
is actually being used instead of the entire buffer.  In addition by
enabling the weak ordering attribute enables a performance improvement for
architectures that can associate a memory ordering with a DMA buffer such
as Sparc.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

On some platforms, syncing a buffer for DMA is expensive. Rather than
sync the whole 2K receive buffer, only synchronise the length of the
frame, which will typically be the MTU, or a much smaller TCP ACK.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch consolidates the code for the ixgbe driver so that it is more
inline with what is already in igb.  The general idea is to just
consolidate functions that represent logical steps in the Rx process so we
can later update them more easily.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Update the driver version to reflect the new devices that it
supports.
Signed-off-by: Mark Rustad <mark.d.rustad@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Since dcbnl_ops is global, it should be prefixed by ixgbe_
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Though not advertised through ethtool, if the link partner advertises a
2.5Gb or 5Gb connection, and the adapter supports it, allow the speed to be
used.
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Krishneil Singh <krishneil.k.singh@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

This reverts commits:

6a254780
9dbbfb0a
40137906

It's too risky to put in this late in the release
cycle.  We'll put these changes into the next merge
window instead.
Signed-off-by: David S. Miller <davem@davemloft.net>

Error reports received from firmware were not being converted from
big endian values, leading to bogus error codes reported on little
endian systems.
Signed-off-by: Thomas Falcon <tlfalcon@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>