As Michal remaind, we should allow to clear ttl inherit. Then we will
have three states:

1. set the flag, and do ttl inherit.
2. do not set the flag, use configured ttl value, or default ttl (0) if
   not set.
3. disable ttl inherit, use previous configured ttl value, or default ttl (0).

Fixes: 52d0d404 ("geneve: add ttl inherit support")
CC: Michal Kubecek <mkubecek@suse.cz>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This traffic scheduler allows traffic classes states (transmission
allowed/not allowed, in the simplest case) to be scheduled, according
to a pre-generated time sequence. This is the basis of the IEEE
802.1Qbv specification.

Example configuration:

tc qdisc replace dev enp3s0 parent root handle 100 taprio \
          num_tc 3 \
	  map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 \
	  queues 1@0 1@1 2@2 \
	  base-time 1528743495910289987 \
	  sched-entry S 01 300000 \
	  sched-entry S 02 300000 \
	  sched-entry S 04 300000 \
	  clockid CLOCK_TAI

The configuration format is similar to mqprio. The main difference is
the presence of a schedule, built by multiple "sched-entry"
definitions, each entry has the following format:

     sched-entry <CMD> <GATE MASK> <INTERVAL>

The only supported <CMD> is "S", which means "SetGateStates",
following the IEEE 802.1Qbv-2015 definition (Table 8-6). <GATE MASK>
is a bitmask where each bit is a associated with a traffic class, so
bit 0 (the least significant bit) being "on" means that traffic class
0 is "active" for that schedule entry. <INTERVAL> is a time duration
in nanoseconds that specifies for how long that state defined by <CMD>
and <GATE MASK> should be held before moving to the next entry.

This schedule is circular, that is, after the last entry is executed
it starts from the first one, indefinitely.

The other parameters can be defined as follows:

 - base-time: specifies the instant when the schedule starts, if
  'base-time' is a time in the past, the schedule will start at

 	      base-time + (N * cycle-time)

   where N is the smallest integer so the resulting time is greater
   than "now", and "cycle-time" is the sum of all the intervals of the
   entries in the schedule;

 - clockid: specifies the reference clock to be used;

The parameters should be similar to what the IEEE 802.1Q family of
specification defines.
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Vasundhara Volam says:

====================
bnxt_en: devlink param updates

This patchset adds support for 3 generic and 1 driver-specific devlink
parameters. Add documentation for these configuration parameters.

Also, this patchset adds support to return proper error code if
HWRM_NVM_GET/SET_VARIABLE commands return error code
HWRM_ERR_CODE_RESOURCE_ACCESS_DENIED.

v3->v4:
-Remove extra definition of NVM_OFF_HW_TC_OFFLOAD from bnxt_devlink.h
-Remove type information for generic parameters from
devlink-params-bnxt.txt

v2->v3:
-Remove description of generic parameters from devlink-params-bnxt.txt

v1->v2:
-Remove hw_tc_offload parameter.
-Update all patches with Cc of MAINTAINERS.
-Add more description in commit message for device specific parameter.
-Add a new Documentation/networking/devlink-params.txt with some
generic devlink parameters information.
-Add a new Documentation/networking/devlink-params-bnxt.txt with devlink
parameters information that are supported by bnxt_en driver.
====================
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This patch adds a new file to add information about configuration
parameters that are supported by bnxt_en driver via devlink.

Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: linux-doc@vger.kernel.org
Cc: Jiri Pirko <jiri@mellanox.com>
Cc: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This patch adds a new file to add information about some of the
generic configuration parameters set via devlink.

Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: linux-doc@vger.kernel.org
Cc: Jiri Pirko <jiri@mellanox.com>
Cc: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This patch adds following driver-specific permanent mode boolean
parameter.

gre_ver_check - Generic Routing Encapsulation(GRE) version check
will be enabled in the device. If disabled, device skips version
checking for GRE packets.

Cc: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This patch adds support for following generic permanent mode
devlink parameters. They can be modified using devlink param
commands.

msix_vec_per_pf_max - This param sets the number of MSIX vectors
that the device requests from the host on driver initialization.
This value is set in the device which limits MSIX vectors per PF.

msix_vec_per_pf_min - This param sets the number of minimal MSIX
vectors required for the device initialization. Value 0 indicates
a default value is selected. This value is set in the device which
limits MSIX vectors per PF.

Cc: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Return proper error code when Firmware returns
HWRM_ERR_CODE_RESOURCE_ACCESS_DENIED for HWRM_NVM_GET/SET_VARIABLE
commands.

Cc: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This patch adds support for ignore_ari generic permanent mode
devlink parameter. This parameter is disabled by default. It can be
enabled using devlink param commands.

ignore_ari - If enabled, device ignores ARI(Alternate Routing ID)
capability, even when platforms has the support and creates same number
of partitions when platform does not support ARI capability.

Cc: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

msix_vec_per_pf_min - This param sets the number of minimal MSIX
vectors required for the device initialization. This value is set
in the device which limits MSIX vectors per PF.

Cc: Jiri Pirko <jiri@mellanox.com>
Cc: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

msix_vec_per_pf_max - This param sets the number of MSIX vectors
that the device requests from the host on driver initialization.
This value is set in the device which is applicable per PF.

Cc: Jiri Pirko <jiri@mellanox.com>
Cc: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

ignore_ari - Device ignores ARI(Alternate Routing ID) capability,
even when platforms has the support and creates same number of
partitions when platform does not support ARI capability.

Cc: Jiri Pirko <jiri@mellanox.com>
Cc: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Clang warns when one enumerated type is implicitly converted to another.

drivers/net/ethernet/qlogic/qed/qed_ll2.c:799:32: warning: implicit
conversion from enumeration type 'enum core_tx_dest' to different
enumeration type 'enum qed_ll2_tx_dest' [-Wenum-conversion]
                tx_pkt.tx_dest = p_ll2_conn->tx_dest;
                               ~ ~~~~~~~~~~~~^~~~~~~
1 warning generated.

Fix this by using a switch statement to convert between the enumerated
values since they are not 1 to 1, which matches how the rest of the
driver handles this conversion.

Link: https://github.com/ClangBuiltLinux/linux/issues/125Suggested-by: Tomer Tayar <Tomer.Tayar@cavium.com>
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Acked-by: Tomer Tayar <Tomer.Tayar@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Saeed Mahameed says:

====================
mlx5-updates-2018-10-03

mlx5 core driver and ethernet netdev updates, please note there is a small
devlink releated update to allow extack argument to eswitch operations.

From Eli Britstein,
1) devlink: Add extack argument to the eswitch related operations
2) net/mlx5e: E-Switch, return extack messages for failures in the e-switch devlink callbacks
3) net/mlx5e: Add extack messages for TC offload failures

From Eran Ben Elisha,
4) mlx5e: Add counter for aRFS rule insertion failures

From Feras Daoud
5) Fast teardown support for mlx5 device
This change introduces the enhanced version of the "Force teardown" that
allows SW to perform teardown in a faster way without the need to reclaim
all the FW pages.
Fast teardown provides the following advantages:
    1- Fix a FW race condition that could cause command timeout
    2- Avoid moving to polling mode
    3- Close the vport to prevent PCI ACK to be sent without been scatter
    to memory
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

David Howells says:

====================
rxrpc: Development

Here are some development patches for AF_RXRPC.  The most significant points
are:

 (1) Change the tracepoint that indicates a packet has been transmitted
     into one that indicates a packet is about to be transmitted.  Without
     this, the response tracepoint may occur first if the round trip is
     fast enough.

 (2) Sort out AFS address list handling to better enforce maximum capacity
     to use helper functions to fill them and to do an insertion sort to
     order them.  This is here to make (3) easier.

 (3) Keep AF_INET addresses as AF_INET addresses rather than converting
     them to AF_INET6 in both AF_RXRPC and kAFS.  I hadn't realised that a
     UDP6 socket would just call down into UDP4 if given an AF_INET
     address.

 (4) Allow the timestamp on the first DATA packet of a reply to be
     retrieved by a kernel service.  This will give the kAFS a more
     accurate base from which to calculate the callback promise expiration.

 (5) Allow the rxrpc protocol epoch value to be retrieved from an incoming
     call.  This will allow kAFS to determine if the fileserver restarted
     and if two addresses apparently assigned to the same fileserver
     actually are different boxes.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Allow the DNS resolver to retrieve a set of servers and their associated
addresses, ports, preference and weight ratings.

In terms of communication with userspace, "srv=1" is added to the callout
string (the '1' indicating the maximum data version supported by the
kernel) to ask the userspace side for this.

If the userspace side doesn't recognise it, it will ignore the option and
return the usual text address list.

If the userspace side does recognise it, it will return some binary data
that begins with a zero byte that would cause the string parsers to give an
error.  The second byte contains the version of the data in the blob (this
may be between 1 and the version specified in the callout data).  The
remainder of the payload is version-specific.

In version 1, the payload looks like (note that this is packed):

	u8	Non-string marker (ie. 0)
	u8	Content (0 => Server list)
	u8	Version (ie. 1)
	u8	Source (eg. DNS_RECORD_FROM_DNS_SRV)
	u8	Status (eg. DNS_LOOKUP_GOOD)
	u8	Number of servers
	foreach-server {
		u16	Name length (LE)
		u16	Priority (as per SRV record) (LE)
		u16	Weight (as per SRV record) (LE)
		u16	Port (LE)
		u8	Source (eg. DNS_RECORD_FROM_NSS)
		u8	Status (eg. DNS_LOOKUP_GOT_NOT_FOUND)
		u8	Protocol (eg. DNS_SERVER_PROTOCOL_UDP)
		u8	Number of addresses
		char[]	Name (not NUL-terminated)
		foreach-address {
			u8		Family (AF_INET{,6})
			union {
				u8[4]	ipv4_addr
				u8[16]	ipv6_addr
			}
		}
	}

This can then be used to fetch a whole cell's VL-server configuration for
AFS, for example.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Trivial fix to spelling mistakes in dev_dbg warning messages

"Reloade" -> "Reload"
"chang" -> "change"
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Merge branch 'ieee802154-for-davem-2018-10-04' of git://git.kernel.org/pub/scm/linux/kernel/git/sschmidt/wpan-next

Stefan Schmidt says:

====================
pull-request: ieee802154-next 2018-10-04

An update from ieee802154 for *net-next*

A very quite cycle in the ieee802154 subsystem. We only have two cleanup
patches for this pull request.
Xue removed the platform_data struct handling from the mcr20a driver and
Alexander cleaned up some left overs in the hwsim driver.

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

Allow the epoch value to be queried on a server connection.  This is in the
rxrpc header of every packet for use in routing and is derived from the
client's state.  It's also not supposed to change unless the client gets
restarted.

AFS can make use of this information to deduce whether a fileserver has
been restarted because the fileserver makes client calls to the filesystem
driver's cache manager to send notifications (ie. callback breaks) about
conflicting changes from other clients.  These convey the fileserver's own
epoch value back to the filesystem.
Signed-off-by: David Howells <dhowells@redhat.com>

Allow the timestamp on the sk_buff holding the first DATA packet of a reply
to be queried.  This can then be used as a base for the expiry time
calculation on the callback promise duration indicated by an operation
result.
Signed-off-by: David Howells <dhowells@redhat.com>

rxrpc_extract_addr_from_skb() doesn't use the argument that points to the
local endpoint, so remove the argument.
Signed-off-by: David Howells <dhowells@redhat.com>

AF_RXRPC opens an IPv6 socket through which to send and receive network
packets, both IPv6 and IPv4.  It currently turns AF_INET addresses into
AF_INET-as-AF_INET6 addresses based on an assumption that this was
necessary; on further inspection of the code, however, it turns out that
the IPv6 code just farms packets aimed at AF_INET addresses out to the IPv4
code.

Fix AF_RXRPC to use AF_INET addresses directly when given them.

Fixes: 7b674e39 ("rxrpc: Fix IPv6 support")
Signed-off-by: David Howells <dhowells@redhat.com>

Sort address lists so that they are in logical ascending order rather than
being partially in ascending order of the BE representations of those
values.
Signed-off-by: David Howells <dhowells@redhat.com>

Make the address list string parser use the helper functions for adding
addresses to an address list so that they end up appropriately sorted.
This will better handles overruns and make them easier to compare.

It also reduces the number of places that addresses are handled, making it
easier to fix the handling.
Signed-off-by: David Howells <dhowells@redhat.com>

Note the maximum allocated capacity in an afs_addr_list struct and discard
addresses that would exceed it in afs_merge_fs_addr{4,6}().

Also, since the current maximum capacity is less than 255, reduce the
relevant members to bytes.
Signed-off-by: David Howells <dhowells@redhat.com>

Print the data Tx trace line before transmitting so that it appears before
the trace lines indicating success or failure of the transmission.  This
makes the trace log less confusing.
Signed-off-by: David Howells <dhowells@redhat.com>

rxrpc_lose_skb() is now exactly the same as rxrpc_free_skb(), so remove it
and use the latter instead.
Signed-off-by: David Howells <dhowells@redhat.com>

Minor conflict in net/core/rtnetlink.c, David Ahern's bug fix in 'net'
overlapped the renaming of a netlink attribute in net-next.
Signed-off-by: David S. Miller <davem@davemloft.net>

Today mlx5 devices support two teardown modes:
1- Regular teardown
2- Force teardown

This change introduces the enhanced version of the "Force teardown" that
allows SW to perform teardown in a faster way without the need to reclaim
all the pages.

Fast teardown provides the following advantages:
1- Fix a FW race condition that could cause command timeout
2- Avoid moving to polling mode
3- Close the vport to prevent PCI ACK to be sent without been scatter
to memory
Signed-off-by: Feras Daoud <ferasda@mellanox.com>
Reviewed-by: Majd Dibbiny <majd@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Count aRFS rules insertion failure for ethtool output. In addition, move
the error print into debug prints mechanism, as it could flood the dmesg
and reduce system BW dramatically.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Return tc extack messages for failures to user space.
Messages provide reasons for not being able to offload rules to HW.
Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Return extack messages for failures in the e-switch devlink callbacks.
Messages provide reasons for not being able to issue the operation.
Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Add extack argument to the eswitch related operations.
Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

David writes:
  "Networking fixes:
   1) Prefix length validation in xfrm layer, from Steffen Klassert.

   2) TX status reporting fix in mac80211, from Andrei Otcheretianski.

   3) Fix hangs due to TX_DROP in mac80211, from Bob Copeland.

   4) Fix DMA error regression in b43, from Larry Finger.

   5) Add input validation to xenvif_set_hash_mapping(), from Jan Beulich.

   6) SMMU unmapping fix in hns driver, from Yunsheng Lin.

   7) Bluetooh crash in unpairing on SMP, from Matias Karhumaa.

   8) WoL handling fixes in the phy layer, from Heiner Kallweit.

   9) Fix deadlock in bonding, from Mahesh Bandewar.

   10) Fill ttl inherit infor in vxlan driver, from Hangbin Liu.

   11) Fix TX timeouts during netpoll, from Michael Chan.

   12) RXRPC layer fixes from David Howells.

   13) Another batch of ndo_poll_controller() removals to deal with
       excessive resource consumption during load.  From Eric Dumazet.

   14) Fix a specific TIPC failure secnario, from LUU Duc Canh.

   15) Really disable clocks in r8169 during suspend so that low
       power states can actually be reached.

   16) Fix SYN backlog lockdep issue in tcp and dccp, from Eric Dumazet.

   17) Fix RCU locking in netpoll SKB send, which shows up in bonding,
       from Dave Jones.

   18) Fix TX stalls in r8169, from Heiner Kallweit.

   19) Fix locksup in nfp due to control message storms, from Jakub
       Kicinski.

   20) Various rmnet bug fixes from Subash Abhinov Kasiviswanathan and
       Sean Tranchetti.

   21) Fix use after free in ip_cmsg_recv_dstaddr(), from Eric Dumazet."

* gitolite.kernel.org:/pub/scm/linux/kernel/git/davem/net: (122 commits)
  ixgbe: check return value of napi_complete_done()
  sctp: fix fall-through annotation
  r8169: always autoneg on resume
  ipv4: fix use-after-free in ip_cmsg_recv_dstaddr()
  net: qualcomm: rmnet: Fix incorrect allocation flag in receive path
  net: qualcomm: rmnet: Fix incorrect allocation flag in transmit
  net: qualcomm: rmnet: Skip processing loopback packets
  net: systemport: Fix wake-up interrupt race during resume
  rtnl: limit IFLA_NUM_TX_QUEUES and IFLA_NUM_RX_QUEUES to 4096
  bonding: fix warning message
  inet: make sure to grab rcu_read_lock before using ireq->ireq_opt
  nfp: avoid soft lockups under control message storm
  declance: Fix continuation with the adapter identification message
  net: fec: fix rare tx timeout
  r8169: fix network stalls due to missing bit TXCFG_AUTO_FIFO
  tun: napi flags belong to tfile
  tun: initialize napi_mutex unconditionally
  tun: remove unused parameters
  bond: take rcu lock in netpoll_send_skb_on_dev
  rtnetlink: Fail dump if target netnsid is invalid
  ...

Jeff Kirsher says:

====================
10GbE Intel Wired LAN Driver Updates 2018-10-03

This series contains updates to ixgbe/ixgbevf and few fixes for i40e & iavf.

Shannon Nelson fixes the message length for IPsec mailbox messages.

Radoslaw fixes a transmit hang that occurs when XDP_TX exceeds the queue
limit.  Fixes a crash when we restor flow director filters after a reset.

YueHaibing cleans up dead code, which did not have any callers.

Dan Carpenter fixes an "off by one" error in IPsec for ixgbe.

Nathan Chancellor fixes the i40e driver to use the correct enum for link
speed.  Also remove a debug statement since it was not producing useful
information and equated to always "TRUE".

Most notably, Björn introduces zero-copy AF_XDP support for the ixgbe
driver.  The ixgbe zero-copy code is located in its own file ixgbe_xsk.[ch],
analogous to the i40e ZC support. Again, as in i40e, code paths have
been copied from the XDP path to the zero-copy path. Going forward we
will try to generalize more code between the AF_XDP ZC drivers, and
also reduce the heavy C&P.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

The NIC driver should only enable interrupts when napi_complete_done()
returns true. This patch adds the check for ixgbe.

Cc: stable@vger.kernel.org # 4.10+
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Song Liu <songliubraving@fb.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

This trivial patch fixes a typo in iavf.h.
Signed-off-by: Rami Rosen <ramirose@gmail.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch adds zero-copy Tx support for AF_XDP sockets. It implements
the ndo_xsk_async_xmit netdev ndo and performs all the Tx logic from a
NAPI context. This means pulling egress packets from the Tx ring,
placing the frames on the NIC HW descriptor ring and completing sent
frames back to the application via the completion ring.

The regular XDP Tx ring is used for AF_XDP as well. This rationale for
this is as follows: XDP_REDIRECT guarantees mutual exclusion between
different NAPI contexts based on CPU id. In other words, a netdev can
XDP_REDIRECT to another netdev with a different NAPI context, since
the operation is bound to a specific core and each core has its own
hardware ring.

As the AF_XDP Tx action is running in the same NAPI context and using
the same ring, it will also be protected from XDP_REDIRECT actions
with the exact same mechanism.

As with AF_XDP Rx, all AF_XDP Tx specific functions are added to
ixgbe_xsk.c.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Tested-by: William Tu <u9012063@gmail.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch prepares for the upcoming zero-copy Tx functionality by
moving common functions used both by the regular path and zero-copy
path.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Tested-by: William Tu <u9012063@gmail.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch adds zero-copy Rx support for AF_XDP sockets. Instead of
allocating buffers of type MEM_TYPE_PAGE_SHARED, the Rx frames are
allocated as MEM_TYPE_ZERO_COPY when AF_XDP is enabled for a certain
queue.

All AF_XDP specific functions are added to a new file, ixgbe_xsk.c.

Note that when AF_XDP zero-copy is enabled, the XDP action XDP_PASS
will allocate a new buffer and copy the zero-copy frame prior passing
it to the kernel stack.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Tested-by: William Tu <u9012063@gmail.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>