Since commit f7f37e7f ("ixgbe: handle close/suspend race with
netif_device_detach/present") ixgbe_close_suspend is called, from
ixgbe_close, only if the device is present, i.e. if it isn't detached.
That exposed a situation where IRQs weren't freed if a PCI error
recovery system opts to remove the device. For such case the pci channel
state is set to pci_channel_io_perm_failure and ixgbe_io_error_detected
was returning PCI_ERS_RESULT_DISCONNECT before calling
ixgbe_close_suspend consequentially not freeing IRQ and crashing when
the remove handler calls pci_disable_device, hitting a BUG_ON at
free_msi_irqs, which asserts that there is no non-free IRQ associated
with the device to be removed:

BUG_ON(irq_has_action(entry->irq + i));

The issue is fixed by calling the ixgbe_close_suspend before evaluate
the pci channel state.
Reported-by: NNaresh Bannoth <nbannoth@in.ibm.com>
Reported-by: NAbdul Haleem <abdhalee@in.ibm.com>
Signed-off-by: NMauro S M Rodrigues <maurosr@linux.vnet.ibm.com>
Reviewed-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Sparse complains valid conversions between restricted types, force
attribute is used to avoid those warnings.
Signed-off-by: NCathy Zhou <cathy.zhou@oracle.com>
Reviewed-by: NShannon Nelson <shannon.nelson@oracle.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

After many years of having a ~30 line copyright and license header to our
source files, we are finally able to reduce that to one line with the
advent of the SPDX identifier.

Also caught a few files missing the SPDX license identifier, so fixed
them up.
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Acked-by: NShannon Nelson <shannon.nelson@oracle.com>
Acked-by: NRichard Cochran <richardcochran@gmail.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The original implementation for macvlan offload has us performing a full
port reset every time we added a new macvlan. This shouldn't be necessary
and can be avoided with a few behavior changes.

This patches updates the logic for the queues so that we have essentially 3
possible configurations for macvlan offload. They consist of 15 macvlans
with 4 queues per macvlan, 31 macvlans with 2 queues per macvlan, and 63
macvlans with 1 queue per macvlan. As macvlans are added you will encounter
up to 3 total resets if you add all the way up to 63, and after that the
device will stay in the mode supporting up to 63 macvlans until the L2FW
flag is cleared.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch drops the real_adapter member from the fwd_adapter structure.
The general idea behind the change is that the real_adapter is carrying
unnecessary data since we could always just grab the adapter structure
from netdev_priv(macvlan->lowerdev) if we really needed to get at it.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Both the ixgbe and fm10k drivers support destination filtering.

Instead of adding a ton of complexity to support either source or passthru
mode we can instead just avoid offloading them for now. Doing this we avoid
leaking packets into interfaces that aren't meant to receive them.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Drop dead code now that we shouldn't be receiving broadcast or multicast
frames on the queues associated to the macvlan netdev.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change makes it so that we use a software path for packets that are
going to be locally switched between two macvlan interfaces on the same
device. In addition we resort to software replication of broadcast and
multicast packets instead of offloading that to hardware.

The general idea is that using the device for east/west traffic local to
the system is extremely inefficient. We can only support up to whatever the
PCIe limit is for any given device so this caps us at somewhere around 20G
for devices supported by ixgbe. This is compounded even further when you
take broadcast and multicast into account as a single 10G port can come to
a crawl as a packet is replicated up to 60+ times in some cases. In order
to get away from that I am implementing changes so that we handle
broadcast/multicast replication and east/west local traffic all in
software.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change renames the fwd_priv member to accel_priv as this more
accurately reflects the actual purpose of this value. In addition I am
adding an accessor which will allow us to further abstract this in the
future if needed.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Drop the code for handling macvlan specific unicast lists. It isn't needed
since we don't take any efforts to maintain it when we bring the interface
up and it takes the slow path anyway.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Changing API ndo_xdp_xmit to take a struct xdp_frame instead of struct
xdp_buff.  This brings xdp_return_frame and ndp_xdp_xmit in sync.

This builds towards changing the API further to become a bulk API,
because xdp_buff is not a queue-able object while xdp_frame is.

V4: Adjust for commit 59655a5b ("tuntap: XDP_TX can use native XDP")
V7: Adjust for commit d9314c47 ("i40e: add support for XDP_REDIRECT")
Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Changing API xdp_return_frame() to take struct xdp_frame as argument,
seems like a natural choice. But there are some subtle performance
details here that needs extra care, which is a deliberate choice.

When de-referencing xdp_frame on a remote CPU during DMA-TX
completion, result in the cache-line is change to "Shared"
state. Later when the page is reused for RX, then this xdp_frame
cache-line is written, which change the state to "Modified".

This situation already happens (naturally) for, virtio_net, tun and
cpumap as the xdp_frame pointer is the queued object.  In tun and
cpumap, the ptr_ring is used for efficiently transferring cache-lines
(with pointers) between CPUs. Thus, the only option is to
de-referencing xdp_frame.

It is only the ixgbe driver that had an optimization, in which it can
avoid doing the de-reference of xdp_frame.  The driver already have
TX-ring queue, which (in case of remote DMA-TX completion) have to be
transferred between CPUs anyhow.  In this data area, we stored a
struct xdp_mem_info and a data pointer, which allowed us to avoid
de-referencing xdp_frame.

To compensate for this, a prefetchw is used for telling the cache
coherency protocol about our access pattern.  My benchmarks show that
this prefetchw is enough to compensate the ixgbe driver.

V7: Adjust for commit d9314c47 ("i40e: add support for XDP_REDIRECT")
V8: Adjust for commit bd658dda ("net/mlx5e: Separate dma base address
and offset in dma_sync call")
Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use the IDA infrastructure for getting a cyclic increasing ID number,
that is used for keeping track of each registered allocator per
RX-queue xdp_rxq_info.  Instead of using the IDR infrastructure, which
uses a radix tree, use a dynamic rhashtable, for creating ID to
pointer lookup table, because this is faster.

The problem that is being solved here is that, the xdp_rxq_info
pointer (stored in xdp_buff) cannot be used directly, as the
guaranteed lifetime is too short.  The info is needed on a
(potentially) remote CPU during DMA-TX completion time . In an
xdp_frame the xdp_mem_info is stored, when it got converted from an
xdp_buff, which is sufficient for the simple page refcnt based recycle
schemes.

For more advanced allocators there is a need to store a pointer to the
registered allocator.  Thus, there is a need to guard the lifetime or
validity of the allocator pointer, which is done through this
rhashtable ID map to pointer. The removal and validity of of the
allocator and helper struct xdp_mem_allocator is guarded by RCU.  The
allocator will be created by the driver, and registered with
xdp_rxq_info_reg_mem_model().

It is up-to debate who is responsible for freeing the allocator
pointer or invoking the allocator destructor function.  In any case,
this must happen via RCU freeing.

Use the IDA infrastructure for getting a cyclic increasing ID number,
that is used for keeping track of each registered allocator per
RX-queue xdp_rxq_info.

V4: Per req of Jason Wang
- Use xdp_rxq_info_reg_mem_model() in all drivers implementing
  XDP_REDIRECT, even-though it's not strictly necessary when
  allocator==NULL for type MEM_TYPE_PAGE_SHARED (given it's zero).

V6: Per req of Alex Duyck
- Introduce rhashtable_lookup() call in later patch

V8: Address sparse should be static warnings (from kbuild test robot)
Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Extend struct ixgbe_tx_buffer to store the xdp_mem_info.

Notice that this could be optimized further by putting this into
a union in the struct ixgbe_tx_buffer, but this patchset
works towards removing this again.  Thus, this is not done.
Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The current page counting scheme assumes that the reference count
cannot decrease until the received frame is sent to the upper layers
of the networking stack. This assumption does not hold for the
XDP_REDIRECT action, since a page (pointed out by xdp_buff) can have
its reference count decreased via the xdp_do_redirect call.

To work around that, we now start off by a large page count and then
don't allow a refcount less than two.
Signed-off-by: NBjörn Töpel <bjorn.topel@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Fix things up to support TSO offload in conjunction
with IPsec hw offload.  This raises throughput with
IPsec offload on to nearly line rate.
Signed-off-by: NShannon Nelson <shannon.nelson@oracle.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Since the ipsec data fields will be zero anyway in the non-ipsec
case, we can remove the conditional jump.
Suggested-by: NAlexander Duyck <alexander.duyck@gmail.com>
Signed-off-by: NShannon Nelson <shannon.nelson@oracle.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

With the patch
commit f8aa2696b4af ("esp: check the NETIF_F_HW_ESP_TX_CSUM bit before segmenting")
we no longer need to protect ourself from checksum
offload requests on IPsec packets, so we can remove
the check in our .ndo_features_check callback.
Signed-off-by: NShannon Nelson <shannon.nelson@oracle.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Add status register reads and delay between reads to ixgbe_check_remove.
Registers can read 0xFFFFFFFF during PCI reset, which causes the driver
to remove the adapter. The additional status register reads can reduce the
chance of this race condition.

If the status register is not 0xFFFFFFFF, then ixgbe_check_remove returns
the value of the register being read.
Signed-off-by: NPaul Greenwalt <paul.greenwalt@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Add check for build_skb enabled ring in ixgbe_dma_sync_frag().
In that case &skb_shinfo(skb)->frags[0] may not always be set which
can lead to a crash. Instead we derive the page offset from skb->data.

Fixes: 42073d91
("ixgbe: Have the CPU take ownership of the buffers sooner")
CC: stable <stable@vger.kernel.org>
Reported-by: NAmbarish Soman <asoman@redhat.com>
Suggested-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

On hardware which supports timestamping all packets, the timestamps are
recorded in the packet buffer, and the driver no longer uses or reads
the registers. This makes the logic for checking and clearing Rx
timestamp hangs meaningless.

If we run the ixgbe_ptp_rx_hang() function in this case, then the driver
will continuously spam the log output with "Clearing Rx timestamp hang".
These messages are spurious, and confusing to end users.

The original code in commit a9763f3c ("ixgbe: Update PTP to support
X550EM_x devices", 2015-12-03) did have a flag PTP_RX_TIMESTAMP_IN_REGISTER
which was intended to be used to avoid the Rx timestamp hang check,
however it did not actually check the flag before calling the function.

Do so now in order to stop the checks and prevent the spurious log
messages.

Fixes: a9763f3c ("ixgbe: Update PTP to support X550EM_x devices", 2015-12-03)
Signed-off-by: NJacob Keller <jacob.e.keller@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

commit 2de6aa3a ("ixgbe: Add support for padding packet")

Uses RXDCTL.RLPML to limit the maximum frame size on Rx when using
build_skb. Unfortunately that register does not work on 82599.

Added an explicit check to avoid setting this register on 82599 MAC.

Extended the comment related to the setting of RXDCTL.RLPML to better
explain its purpose.
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Since 5G link speed is supported by some devices, add reporting of 5G link
speed.
Signed-off-by: NPaul Greenwalt <paul.greenwalt@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

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>

With all the support code in place we can now link in the ipsec
offload operations and set the ESP feature flag for the XFRM
subsystem to see.
Signed-off-by: NShannon Nelson <shannon.nelson@oracle.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Add a simple statistic to count the ipsec offloads.
Signed-off-by: NShannon Nelson <shannon.nelson@oracle.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

If the skb has a security association referenced in the skb, then
set up the Tx descriptor with the ipsec offload bits.  While we're
here, we fix an oddly named field in the context descriptor struct.
Signed-off-by: NShannon Nelson <shannon.nelson@oracle.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

If the chip sees and decrypts an ipsec offload, set up the skb
sp pointer with the ralated SA info.  Since the chip is rude
enough to keep to itself the table index it used for the
decryption, we have to do our own table lookup, using the
hash for speed.
Signed-off-by: NShannon Nelson <shannon.nelson@oracle.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

On a chip reset most of the table contents are lost, so must be
restored.  This scans the driver's ipsec tables and restores both
the filled and empty table slots to their pre-reset values.
Signed-off-by: NShannon Nelson <shannon.nelson@oracle.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Add the functions for setting up and removing offloaded SAs (Security
Associations) with the x540 hardware.  We set up the callback structure
but we don't yet set the hardware feature bit to be sure the XFRM service
won't actually try to use us for an offload yet.

The software tables are made up to mimic the hardware tables to make it
easier to track what's in the hardware, and the SA table index is used
for the XFRM offload handle.  However, there is a hashing field in the
Rx SA tracking that will be used to facilitate faster table searches in
the Rx fast path.
Signed-off-by: NShannon Nelson <shannon.nelson@oracle.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Add a few routines to make access to the ipsec registers just a little
easier, and throw in the beginnings of an initialization.
Signed-off-by: NShannon Nelson <shannon.nelson@oracle.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Recent checks added for formatting kernel-doc comments are causing warnings
if W= is run with a non-zero value.  This patch fixes function comments to
resolve warnings when W=1 is used.
Signed-off-by: NTony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This update makes it so that we report the actual number of Tx queues via
real_num_tx_queues but are still restricted to RSS on only the first pool
by setting num_tc equal to 1. Doing this locks us into only having the
ability to setup XPS on the queues in that pool, and only those queues
should be used for transmitting anything other than macvlan traffic.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change makes it so that instead of bringing rings up/down for various
we just update the netdev pointer for the Rx ring and set or clear the MAC
filter for the interface. By doing it this way we can avoid a number of
races and issues in the code as things were getting messy with the macvlan
clean-up racing with the interface clean-up to bring the rings down on
shutdown.

With this change we opt to leave the rings owned by the PF interface for
both Tx and Rx and just direct the packets once they are received to the
macvlan netdev.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

We should not be stopping/starting the upper devices Tx queues when
handling a macvlan offload. Instead we should be stopping and starting
traffic on our own queues.

In order to prevent us from doing this I am updating the code so that we no
longer change the queue configuration on the upper device, nor do we update
the queue_index on our own device. Instead we can just use the queue index
for our local device and not update the netdev in the case of the transmit
rings.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

We shouldn't be recording the Rx queue on macvlan offloaded frames since
the macvlan is normally brought up as a single queue device, and it will
trigger warnings for RPS if we have recorded queue IDs larger than the
"real_num_rx_queues" value recorded for the device.

Instead we should be recording the macvlan statistics since we are
bypassing the normal macvlan statistics that would have been generated by
the receive path.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

The code throughout ixgbe was assuming that dev->num_tc was populated and
configured with the driver, when in fact this can be configured via mqprio
without any hardware coordination other than restricting us to the real
number of Tx queues we advertise.

Instead of handling things this way we need to keep a local copy of the
number of TCs in use so that we don't accidentally pull in the TC
configuration from mqprio when it is configured in software mode.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

We might as well configure the limit to default to 1 pool always for the
interface. This accounts for the fact that the PF counts as 1 pool if
SR-IOV is enabled, and in general we are always running in 1 pool mode when
RSS or DCB is enabled as well, though we don't need to actually evaluate
any of the VMDq features in those cases.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

The macvlan driver itself will validate the MAC address that is configured
for a given interface. There is no need for us to verify it again.

Instead we should be checking to verify that we actually allocate the filter
and have not run out of resources to configure a MAC rule in our filter
table.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

The l2 acceleration private pointer isn't needed in the ring struct. It
isn't really used anywhere other than to test and see if we are supporting
an offloaded macvlan netdev, and it is much easier to test netdev for not
being ixgbe based to verify that.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>