This reverts commit 068a2de5 (net: release dst entry while
cache-hot for GSO case too)

Before GSO packet segmentation, we already take care of skb->dst if it
can be released.

There is no point adding extra test for every segment in the gso loop.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: Krishna Kumar <krkumar2@in.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, packet_sock has a struct tpacket_stats stats member for
TPACKET_V1 and TPACKET_V2 statistic accounting, and with TPACKET_V3
``union tpacket_stats_u stats_u'' was introduced, where however only
statistics for TPACKET_V3 are held, and when copied to user space,
TPACKET_V3 does some hackery and access also tpacket_stats' stats,
although everything could have been done within the union itself.

Unify accounting within the tpacket_stats_u union so that we can
remove 8 bytes from packet_sock that are there unnecessary. Note that
even if we switch to TPACKET_V3 and would use non mmap(2)ed option,
this still works due to the union with same types + offsets, that are
exposed to the user space.
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There's a 4 byte hole in packet_ring_buffer structure before
prb_bdqc, that can be filled with 'pending' member, thus we can
reduce the overall structure size from 224 bytes to 216 bytes.
This also has the side-effect, that in struct packet_sock 2*4 byte
holes after the embedded packet_ring_buffer members are removed,
and overall, packet_sock can be reduced by 1 cacheline:

Before: size: 1344, cachelines: 21, members: 24
After:  size: 1280, cachelines: 20, members: 24
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Daniel Borkmann says:

====================
This is a joint effort with Willem to bring optional i) tx hw/sw
timestamping into PF_PACKET, that was reported by Paul Chavent,
and ii) to expose the type of timestamp to the user, which is in
the current situation not possible to distinguish with the RX_RING
and TX_RING API (but distinguishable through the normal timestamping
API), reported by Richard Cochran. This set is based on top of
``packet: account statistics only in tpacket_stats_u''. Related
discussion can be found in: http://patchwork.ozlabs.org/patch/238125/
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Bring the timestamping section in sync with the implementation.
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Acked-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, there is no way to find out which timestamp is reported in
tpacket{,2,3}_hdr's tp_sec, tp_{n,u}sec members. It can be one of
SOF_TIMESTAMPING_SYS_HARDWARE, SOF_TIMESTAMPING_RAW_HARDWARE,
SOF_TIMESTAMPING_SOFTWARE, or a fallback variant late call from the
PF_PACKET code in software.

Therefore, report in the tp_status member of the ring buffer which
timestamp has been reported for RX and TX path. This should not break
anything for the following reasons: i) in RX ring path, the user needs
to test for tp_status & TP_STATUS_USER, and later for other flags as
well such as TP_STATUS_VLAN_VALID et al, so adding other flags will
do no harm; ii) in TX ring path, time stamps with PACKET_TIMESTAMP
socketoption are not available resp. had no effect except that the
application setting this is buggy. Next to TP_STATUS_AVAILABLE, the
user also should check for other flags such as TP_STATUS_WRONG_FORMAT
to reclaim frames to the application. Thus, in case TX ts are turned
off (default case), nothing happens to the application logic, and in
case we want to use this new feature, we now can also check which of
the ts source is reported in the status field as provided in the docs.
Reported-by: NRichard Cochran <richardcochran@gmail.com>
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Acked-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This makes it more readable and clearer what bits are still free to
use. The compiler reduces this to a constant for us anyway.
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Acked-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, we only have software timestamping for the TX ring buffer
path, but this limitation stems rather from the implementation. By
just reusing tpacket_get_timestamp(), we can also allow hardware
timestamping just as in the RX path.
Signed-off-by: NDaniel Borkmann <dborkman@redhat.com>
Acked-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When transmit timestamping is enabled at the socket level, record a
timestamp on packets written to a PACKET_TX_RING. Tx timestamps are
always looped to the application over the socket error queue. Software
timestamps are also written back into the packet frame header in the
packet ring.
Reported-by: NPaul Chavent <paul.chavent@onera.fr>
Signed-off-by: NWillem de Bruijn <willemb@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Jeff Kirsher says:

====================
This series contains updates to ixgbe, igb and pci.

The ixgbe changes contains a fix to a possible divide by zero by bailing
out of the ixgbe_update_itr() function if the last interrupt timeslice is
zero.  In addition, support is added for the new OCP x520 adapter as well
as LX support for 82599 devices.  Jacob provides a patch to change
variable wol_supported to wol_enabled to better reflect what the code
is actually doing (i.e. checking if WoL is enabled).

Alex adds SRIOV helper function to pci that will determine if a PF
has any VFs that are currently assigned to a guest.

The remaining 8 patches are against igb and contain the following changes:
* implement SERDES loopback configuration for i210 devices by unsetting
  sigdetect bit, so as to fix Ethtool loopback test failure
* add support for the SMBI semaphore for I210/I211 devices
* implement the new generic pci_vfs_assigned helper function (Alex's PCI
  helper function)
* display warning when link speed is downgraded due to Smartspeed
* ensure that VLAN hardware filtering remains enabled when the device is
  in promiscuous mode and VT mode simultaneously
* cleanup dead code in igb
* bump the driver version

v2: updated the PCI patch to add SRIOV helper function to remove extern
    from the declaration of pci_vfs_assigned in pci.h and return 0 if
    SR-IOV is disabled which is inline with other PCI SR-IOV functions
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Pablo Neira Ayuso says:

====================
The following patchset contains fixes for recently applied
Netfilter/IPVS updates to the net-next tree, most relevantly
they are:

* Fix sparse warnings introduced in the RCU conversion, from
  Julian Anastasov.

* Fix wrong endianness in the size field of IPVS sync messages,
  from Simon Horman.

* Fix missing if checking in nf_xfrm_me_harder, from Dan Carpenter.

* Fix off by one access in the IPVS SCTP tracking code, again from
  Dan Carpenter.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NCarolyn Wyborny <carolyn.wyborny@intel.com>
Tested-by: NAaron Brown <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch removes id defines from the hardware files that will not be
productized for Linux.  These id's were not implemented for support in the
base driver itself, they were just available defines.
Signed-off-by: NCarolyn Wyborny <carolyn.wyborny@intel.com>
Tested-by: NAaron Brown <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

The 82575 manual initialization scripts are not supported on 82580 and
above. Rather than call the function to immediately return, clarify the
code by removing this pointless function call.
Signed-off-by: NMatthew Vick <matthew.vick@intel.com>
Tested-by: NAaron Brown <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

When using the new bridge FDB interface to allow SR-IOV virtual function
network devices to communicate with SW bridged network devices the
physical function is placed into promiscuous mode and hardware VLAN
filtering is disabled.  This defeats the ability to use VLAN tagging
to isolate user networks.  When the device is in promiscuous mode and
VT mode simultaneously ensure that VLAN hardware filtering remains
enabled.
Signed-off-by: NGreg Rose <gregory.v.rose@intel.com>
Tested-by: NSibai Li <sibai.li@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Current igb driver doesn't tell nothing when Link Speed is downgraded due to
SmartSpeed.  As a result, users suspect that there is something wrong with
NIC.  If the cause of it is SmartSpeed, there is no means to replace NIC. This
patch make igb notify users that SmartSpeed worked.
Signed-off-by: NKoki Sanagi <sanagi.koki@jp.fujitsu.com>
Tested-by: NAaron Brown <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change makes it so that the igb driver uses the generic helper
pci_vfs_assigned instead of the igb specific function igb_vfs_are_assigned.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NSibai Li <sibai.li@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This function is meant to add a helper function that will determine if a PF
has any VFs that are currently assigned to a guest.  We currently have been
implementing this function per driver, and going forward I would like to avoid
that by making this function generic and using this helper.

v2: Removed extern from declaration of pci_vfs_assigned in pci.h and return
    0 if SR-IOV is disabled with is inline with other PCI SRIOV functions.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Acked-by: NBjorn Helgaas <bhelgaas@google.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

It was previously thought that, since I210/I211 are single port devices,
they did not need the SMBI semaphore. This is not the case. Add support for
the SMBI semaphore.
Signed-off-by: NMatthew Vick <matthew.vick@intel.com>
Tested-by: NAaron Brown <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch implements SERDES loopback configuration for i210 devices by
unsetting sigdetect bit, so as to fix Ethtool loopback test failure. Old
sigdetect code is also simplified to take care of all devices newer than 82580
Signed-off-by: NAkeem G Abodunrin <akeem.g.abodunrin@intel.com>
Tested-by: NAaron Brown <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch adds a define and WOL support for a new subdevice ID.
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch adds LX support to 82599 devices. This is an alternate patch to
the one suggested by Stefan Behte <s.behte@babiel.com>

In addition this patch includes some cleanups such as:
- removed parenthesis around "x == y ||" lines inside an if statement for
consistency.
- grouped the sx/lx sfp types along with srlr in ixgbe_get_settings() since
they all have the same supported, advertised and port values.
Signed-off-by: NDon Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Reported-by: NStefan Behte <s.behte@babiel.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

The variable wol_supported really is just checking whether it is enabled, rather
than whether it is supported. If it is enabled it will be supported, but this
does not necessarily hold true the other way around. This patch renames the
variable to avoid confusion.
Signed-off-by: NJacob Keller <jacob.e.keller@intel.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch adds support for the new OCP x520 adapter.  This support
includes WoL.
Signed-off-by: NDon Skidmore <donald.c.skidmore@intel.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Protect the code by bailing out of ixgbe_update_itr() when this occurs.
The next call to ixgbe_update_itr will continue to dynamically update ITR.
Signed-of-by: NDon Skidmore <donald.c.skidmore@intel.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This if statement was accidentally dropped in (aaa795ad netfilter:
nat: propagate errors from xfrm_me_harder()) so now it returns
unconditionally.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>

Jitendra Kalsaria says:

====================
This set of patches has following updates:
* Enhanced channel configuration logs by adding logs for various cases.
* Take EPORT out of reset before disabling pause frame generation in the
  adapter.
* Add eSwitch statistics support in ethtool stats.
* Enable interrupt coalescing for 83xx adapter.
* Rename IRQ description.
* Added PCI_VENDOR_ID_QLOGIC, PCI_DEVICE_ID_QLOGIC_834X and PCI_DEVICE_ID_QLOGIC_824X
  for the patch "qlcnic: Add identifying string for 83xx adapter" as per Francois comment.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

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

o Added identifying strings for 8300 Series of adapters.
o updated PCI_VENDOR_ID_QLOGIC and PCI_DEVICE_ID_824X for 8200
  Series adapter.
Signed-off-by: NHimanshu Madhani <himanshu.madhani@qlogic.com>
Signed-off-by: NShahed Shaikh <shahed.shaikh@qlogic.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Here's what modified vectors will look like
in the /proc/interrupts

        MSIx             	INTx
-----------------------------------------
83xx    qlcnic[MB]
	qlcnic-ethX[Rx0]
	qlcnic-ethX[Rx1]
	..
	qlcnic-ethX[RxN]
	qlcnic-ethx[Tx0]        qlcnic[MB+Tx0+Rx0]

82xx    qlcnic-ethX[Rx0]
	qlcnic-ethX[Rx1]
	..
	qlcnic-ethX[Tx0+RxN]    qlcnic-ethX[Tx0+Rx0]
Signed-off-by: NHimanshu Madhani <himanshu.madhani@qlogic.com>
Signed-off-by: NShahed Shaikh <shahed.shaikh@qlogic.com>
Signed-off-by: NJitendra Kalsaria <jitendra.kalsaria@qlogic.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Enable Interrupt coalescing through ethtool on 83xx adapter.
Signed-off-by: NHimanshu Madhani <himanshu.madhani@qlogic.com>
Signed-off-by: NShahed Shaikh <shahed.shaikh@qlogic.com>
Signed-off-by: NJitendra Kalsaria <jitendra.kalsaria@qlogic.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

o Read eSwitch statistics from adapter and display them as part
  of ethtool statistics.
Signed-off-by: NShahed Shaikh <shahed.shaikh@qlogic.com>
Signed-off-by: NJitendra Kalsaria <jitendra.kalsaria@qlogic.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

o Disabling PAUSE requires access to EPORT registers,
  which may cause a wedge, if EPORT is in reset.
Signed-off-by: NManish Chopra <manish.chopra@qlogic.com>
Signed-off-by: NShahed Shaikh <shahed.shaikh@qlogic.com>
Signed-off-by: NJitendra Kalsaria <jitendra.kalsaria@qlogic.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

o Add logs for various failure conditions during channel configuration.
Signed-off-by: NManish Chopra <manish.chopra@qlogic.com>
Signed-off-by: NShahed Shaikh <shahed.shaikh@qlogic.com>
Signed-off-by: NJitendra Kalsaria <jitendra.kalsaria@qlogic.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Yuval Mintz says:

====================
This patch series contains several enhancements, as well as small fixes:

  - Patch [1/5] - Prevent a theoretical problem in our GRO implementation.

  - Patch [2/5] - Support Rx/Tx pause control configuration in autoneg.

  - Patch [3/5] - Enhance support for VF's MAC setting and removal.

  - Patch [4/5] - Fix a small memory leak between bnx2x and cnic.

  - Patch [5/5] - Allow bnx2x to recover after a second slot reset.

Please consider applying these patches to `net-next'.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As part of PCIe Advanced Error Reporting flow, if a fatal PCI error occurs,
the AER driver will cause bnx2x's PCI-core to reset. The driver's PCI error
handlers will in turn restore the PCI configuration space values by calling
`pci_restore_state'.

However, as bnx2x does not save the PCI configuration after restoration,
An additional fatal PCI error will leave the function in an unstable state
until reboot, as the registers in the PCI configuration space will contain
reset values.
Signed-off-by: NYuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: NAriel Elior <ariele@broadcom.com>
Signed-off-by: NEilon Greenstein <eilong@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There exists an `allocation race' between the CNIC and bnx2x drivers,
in which both drivers allocate the same t2 memory while disregarding a possible
previous allocation.

Additionally, due to the current order of memory releases, some of the
ILT memory in the driver is not released correctly when unloading the driver.
Signed-off-by: NYuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: NAriel Elior <ariele@broadcom.com>
Signed-off-by: NEilon Greenstein <eilong@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Improved support for adding/removing vf mac addresses.

This includes the case where HyperVisor forced the address (sampled from
bulletin board), and the case where it did not in which the VF can
configure its own mac address.
Signed-off-by: NDmitry Kravkov <dmitry@broadcom.com>
Signed-off-by: NAriel Elior <ariele@broadcom.com>
Signed-off-by: NYuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: NEilon Greenstein <eilong@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, when link is configured to auto-negotiate the flow control,
disabling RX/TX pause via ethtool doesn't work.

This fixes the behaviour, advertising asymmetric pause in case either one
is exclusively enabled.
Signed-off-by: NYaniv Rosner <yanivr@broadcom.com>
Signed-off-by: NYuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: NEilon Greenstein <eilong@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch introduces a more robust error handling flow in case of incorrect
behaviour by the FW when passing on GRO aggregations.

Although this should never happen (i.e., this is merely a theoretical fix),
if the bnx2x driver was to receive a GRO from FW with protocol other than
IPv4/IPv6, the driver would falsely claim to have performed partial
checksum and set various incorrect fields in the skb header.

Current behaviour of the bnx2x driver (i.e., print an error) is insufficient.
This patch remedies this by simply preventing the false claims.
Signed-off-by: NYuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: NAriel Elior <ariele@broadcom.com>
Signed-off-by: NEilon Greenstein <eilong@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>