In the case of the Tx rings we need to only clear the Tx buffer_info when
we are resetting the rings.  Ideally we do this when we configure the ring
to bring it back up instead of when we are taking it down in order to avoid
dirtying pages we don't need to.

In addition we don't need to clear the Tx descriptor ring since we will
fully repopulate it when we begin transmitting frames and next_to_watch can
be cleared to prevent the ring from being cleaned beyond that point instead
of needing to touch anything in the Tx descriptor ring.

Finally with these changes we can avoid having to reset the skb member of
the Tx buffer_info structure in the cleanup path since the skb will always
be associated with the first buffer which has next_to_watch set.
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NKrishneil Singh <krishneil.k.singh@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Based on commit ec718254
("ixgbe: Improve performance and reduce size of ixgbe_tx_map")

This change is meant to both improve the performance and reduce the size of
ixgbevf_tx_map().

Expand the work done in the main loop by pushing first into tx_buffer.
This allows us to pull in the dma_mapping_error check, the tx_buffer value
assignment, and the initial DMA value assignment to the Tx descriptor.
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NKrishneil Singh <krishneil.k.singh@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Based on commit d2bead57
("igb: Clear Rx buffer_info in configure instead of clean")

This change makes it so that instead of going through the entire ring on Rx
cleanup we only go through the region that was designated to be cleaned up
and stop when we reach the region where new allocations should start.

In addition we can avoid having to perform a memset on the Rx buffer_info
structures until we are about to start using the ring again.
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NKrishneil Singh <krishneil.k.singh@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

We already had placehloders for failed page and buffer allocations.
Added alloc_rx_page and made sure the stats are properly updated and
exposed in ethtool.
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NKrishneil Singh <krishneil.k.singh@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Based on commit bd4171a5
("igb: update code to better handle incrementing page count")

Update the driver code so that we do bulk updates of the page reference
count instead of just incrementing it by one reference at a time.  The
advantage to doing this is that we cut down on atomic operations and
this in turn should give us a slight improvement in cycles per packet.
In addition if we eventually move this over to using build_skb the gains
will be more noticeable.
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NKrishneil Singh <krishneil.k.singh@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Based on commit 5be59554
("igb: update driver to make use of DMA_ATTR_SKIP_CPU_SYNC")
and
commit 7bd17592 ("igb: Add support for DMA_ATTR_WEAK_ORDERING")

Convert the calls to dma_map/unmap_page() to the attributes version
and add DMA_ATTR_SKIP_CPU_SYNC/WEAK_ORDERING which should help
improve performance on some platforms.

Move sync_for_cpu call before we perform a prefetch to avoid
invalidating the first 128 bytes of the packet on architectures where
that call may invalidate the cache.
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NKrishneil Singh <krishneil.k.singh@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Based on:
commit 7ec0116c ("igb: Use length to determine if descriptor is done")

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.

In addition we only reset the Rx descriptor length for descriptor zero when
resetting a ring instead of having to do a memset with 0 over the entire
ring. By doing this we can save some time on initialization.
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NKrishneil Singh <krishneil.k.singh@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Based on commit 64f2525c ("igb: Only DMA sync frame length")

On some architectures synching a buffer for DMA may be expensive.
Instead of the entire 2K receive buffer only synchronize the length of
the frame, which will typically be the MTU or smaller.
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NKrishneil Singh <krishneil.k.singh@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Introduce ixgbevf_can_reuse_page() similar to the change in ixgbe from
commit af43da0d
("ixgbe: Add function for checking to see if we can reuse page")
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NKrishneil Singh <krishneil.k.singh@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Message sends to the local broadcast address (255.255.255.255) require
uc_index or sk_bound_dev_if to be set to an egress device. However,
responses or only received if the socket is bound to the device. This
is overly constraining for processes running in an L3 domain. This
patch allows a socket bound to the VRF device to send to the local
broadcast address by using IP_UNICAST_IF to set the egress interface
with packet receipt handled by the VRF binding.

Similar to IP_MULTICAST_IF, relax the constraint on setting
IP_UNICAST_IF if a socket is bound to an L3 master device. In this
case allow uc_index to be set to an enslaved if sk_bound_dev_if is
an L3 master device and is the master device for the ifindex.

In udp and raw sendmsg, allow uc_index to override the oif if
uc_index master device is oif (ie., the oif is an L3 master and the
index is an L3 slave).
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

William Tu says:

====================
net: erspan: add support for openvswitch

The first patch refactors the erspan header definitions.
Originally, the erspan fields are defined as a group into a __be16 field,
and use mask and offset to access each field.  This is more costly due to
calling ntohs/htons and error-prone.  The first patch changes it to use
bitfields.  The second patch creates erspan.h in UAPI and move the definition
'struct erspan_metadata' to it for later openvswitch to use.  The final patch
introduces the new OVS tunnel key attribute, OVS_TUNNEL_KEY_ATTR_ERSPAN_OPTS,
to program both v1 and v2 erspan tunnel for openvswitch.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The patch adds support for openvswitch to configure erspan
v1 and v2.  The OVS_TUNNEL_KEY_ATTR_ERSPAN_OPTS attr is added
to uapi as a binary blob to support all ERSPAN v1 and v2's
fields.  Note that Previous commit "openvswitch: Add erspan tunnel
support." was reverted since it does not design properly.
Signed-off-by: NWilliam Tu <u9012063@gmail.com>
Acked-by: NPravin B Shelar <pshelar@ovn.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The patch adds a new uapi header file, erspan.h, and moves
the 'struct erspan_metadata' from internal erspan.h to it.
Signed-off-by: NWilliam Tu <u9012063@gmail.com>
Acked-by: NPravin B Shelar <pshelar@ovn.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Originally the erspan fields are defined as a group into a __be16 field,
and use mask and offset to access each field.  This is more costly due to
calling ntohs/htons.  The patch changes it to use bitfields.
Signed-off-by: NWilliam Tu <u9012063@gmail.com>
Acked-by: NPravin B Shelar <pshelar@ovn.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Jakub Kicinski says:

====================
use tc_cls_can_offload_and_chain0() throughout the drivers

This set makes all drivers use a new tc_cls_can_offload_and_chain0()
helper which will set extack in case TC hw offload flag is disabled.

I chose to keep the new helper which also looks at the chain but
renamed it more appropriately.  The rationale being that most drivers
don't accept chains other than 0 and since we have to pass extack
to the helper we can as well pass the entire struct tc_cls_common_offload
and perform the most common checks.

This code makes the assumption that type_data in the callback can
be interpreted as struct tc_cls_common_offload, i.e. the real offload
structure has common part as the first member.  This allows us to
make the check once for all classifier types if driver supports
more than one.

v1:
 - drop the type validation in nfp and netdevsim.
v2:
 - reorder checks in patch 1;
 - split other changes from patch 1;
 - add the i40e patch in;
 - add one more test case - for chain 0 extack.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Make sure netdevsim doesn't allow offload of chains other than 0,
and that it reports the expected extack message.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Drivers should not report errors when offload is not forced.
Check stdout and stderr for familiar messages when with no
skip flags and with skip_hw.  Check for add, replace, and
destroy.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Make use of tc_cls_can_offload_and_chain0() to set extack msg in case
ethtool tc offload flag is not set or chain unsupported.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Make use of tc_cls_can_offload_and_chain0() to set extack msg in case
ethtool tc offload flag is not set or chain unsupported.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Make use of tc_cls_can_offload_and_chain0() to set extack msg in case
ethtool tc offload flag is not set or chain unsupported.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Make use of tc_cls_can_offload_and_chain0() to set extack msg in case
ethtool tc offload flag is not set or chain unsupported.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Make use of tc_cls_can_offload_and_chain0() to set extack msg in case
ethtool tc offload flag is not set or chain unsupported.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NSimon Horman <simon.horman@netronome.com>
Acked-by: NOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Make use of tc_cls_can_offload_and_chain0() to set extack msg in case
ethtool tc offload flag is not set or chain unsupported.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Make use of tc_cls_can_offload_and_chain0() to set extack msg in case
ethtool tc offload flag is not set or chain unsupported.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Make use of tc_cls_can_offload_and_chain0() to set extack msg in case
ethtool tc offload flag is not set or chain unsupported.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Very few (mlxsw) upstream drivers seem to allow offload of chains
other than 0.  Save driver developers typing and add a helper for
checking both if ethtool's TC offload flag is on and if chain is 0.
This helper will set the extack appropriately in both error cases.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NSimon Horman <simon.horman@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Issue found by checkpatch.
Signed-off-by: NRohit Visavalia <rohit.visavalia@softnautics.com>
Acked-by: NMichal Kalderon <michal.kalderon@cavium.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Resolved Warning: networking block comments don't use an empty /* line,
use /* Comment...
Issue found by checkpatch.
Signed-off-by: NRohit Visavalia <rohit.visavalia@softnautics.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Merge branch 'for-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next

Johan Hedberg says:

====================
pull request: bluetooth-next 2018-01-25

Here's one last bluetooth-next pull request for the 4.16 kernel:

 - Improved support for Intel controllers
 - New set_parity method to serdev (agreed with maintainers to be taken
   through bluetooth-next)
 - Fix error path in hci_bcm (missing call to serdev close)
 - New ID for BCM4343A0 UART controller

Please let me know if there are any issues pulling. Thanks.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

t4_wr_mbox_meat_timeout() can be called from both softirq
context and process context, hence protect the mbox with
spin_lock_bh() instead of simple spin_lock()
Signed-off-by: NGanesh Goudar <ganeshgr@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

IPv6 allows routes to be installed when the device is not up (admin up).
Worse, it does not mark it as LINKDOWN. IPv4 does not allow it and really
there is no reason for IPv6 to allow it, so check the flags and deny if
device is admin down.
Signed-off-by: NDavid Ahern <dsahern@gmail.com>
Reviewed-by: NIdo Schimmel <idosch@mellanox.com>
Reviewed-by: NRoopa Prabhu <roopa@cumulusnetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Ursula Braun says:

====================
net/smc: more socket closing improvements

these patches improve the smc behavior for abnormal socket closing.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If a problem for at least one connection of a link group is detected,
the whole link group and all its connections are terminated.
This patch adds a check for healthy link group when trying to reserve
a work request, and checks for healthy connections before starting
a tx worker.
Signed-off-by: NUrsula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If a new connection with a new rmb is added to a link group, its
memory region is registered. If a link group is terminated, a pending
registration requires a wake up.

And consolidate setting of tx_flag peer_conn_abort in smc_lgr_terminate().
Signed-off-by: NUrsula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Once a linkgroup is created successfully, it stays alive for a
certain time to service more connections potentially created.
If one of the initialization steps for a new linkgroup fails,
the linkgroup should not be reused by other connections following.
Signed-off-by: NUrsula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If ib_post_send() fails, terminate all connections of this
link group.
Signed-off-by: NUrsula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

A state transition from closing state SMC_PEERFINCLOSEWAIT to closing
state SMC_APPFINCLOSEWAIT is not allowed. Once a closing indication
from the peer has been received, the socket reaches state SMC_CLOSED.

And receiving a peer_conn_abort just changes the state of the socket
into one of the states SMC_PROCESSABORT or SMC_CLOSED;
sending a peer_conn_abort occurs in smc_close_active() for state
SMC_PROCESSABORT only.
Signed-off-by: NUrsula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If an SMC socket is aborted, the tx worker should be cancelled.
Signed-off-by: NUrsula Braun <ubraun@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Edward Cree says:

====================
sfc: support PTP on 8000 and X2000 series NICs

Starting from the 8000-series (Medford 1), SFC NICs can timestamp TX packets
 sent through an ordinary DMA queue, rather than a special control-plane
 operation as in the 7000-series.  Patches 2-8 implement support for this.
The X2000-series (Medford 2) changes the format of timestamps, from seconds+
 (2^27)ths to seconds + quarter nanoseconds, as well as changing the shift
 of the frequency adjustment for increased precision.  Patches 9-12
 implement support for these changes.
Patch #1 is an unrelated fix for NAPI budget handling, needed in order for
 TX completion changes in the later patches to apply cleanly.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Support increased precision frequency adjustment format (FP44) used
 by Medford2 adapters.
Signed-off-by: NLaurence Evans <levans@solarflare.com>
Signed-off-by: NEdward Cree <ecree@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>