We already force ring sizes to be power of 2 so replace
modulo operations with AND (size - 1) in index calculations.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Introduce a separate buffer allocation function to be called
from NAPI.  We can make assumptions about the context and
buffer size.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Speed up RX processing by moving to the alloc_frag()/build_skb()
paradigm.  Since we're no longer mapping the entire buffer for
DMA add helpers which take care of calculating offsets and
lengths.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

nfp_net_rx() is quite long already and about to get longer.
Move buffer drop/recycle to a helper.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add a helper function to calculate the buffer size at run time.
Buffer lengths will now depend on the FW prepend configuration
instead of assuming the most space consuming configuration and
defaulting to 2k buffers at initialization time.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Don't declare functions as static inline in .c files and
remove dead code it was hiding.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

ether_setup() will be invoked by alloc_etherdev_mqs(), no need
to call it again.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Drop all code related to nfp3200.  It was never widely deployed
as a NIC.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There are few variables in nfp_net_poll() which are used only
once or unused but set.  Remove them.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

et131x: min_mtu 64, max_mtu 9216

altera_tse: min_mtu 64, max_mtu 1500

amd8111e: min_mtu 60, max_mtu 9000

bnad: min_mtu 46, max_mtu 9000

macb: min_mtu 68, max_mtu 1500 or 10240 depending on hardware capability

xgmac: min_mtu 46, max_mtu 9000

cxgb2: min_mtu 68, max_mtu 9582 (pm3393) or 9600 (vsc7326)

enic: min_mtu 68, max_mtu 9000

gianfar: min_mtu 50, max_mu 9586

hns_enet: min_mtu 68, max_mtu 9578 (v1) or 9706 (v2)

ksz884x: min_mtu 60, max_mtu 1894

myri10ge: min_mtu 68, max_mtu 9000

natsemi: min_mtu 64, max_mtu 2024

nfp: min_mtu 68, max_mtu hardware-specific

forcedeth: min_mtu 64, max_mtu 1500 or 9100, depending on hardware

pch_gbe: min_mtu 46, max_mtu 10300

pasemi_mac: min_mtu 64, max_mtu 9000

qcaspi: min_mtu 46, max_mtu 1500
- remove qcaspi_netdev_change_mtu as it is now redundant

rocker: min_mtu 68, max_mtu 9000

sxgbe: min_mtu 68, max_mtu 9000

stmmac: min_mtu 46, max_mtu depends on hardware

tehuti: min_mtu 60, max_mtu 16384
- driver had no max mtu checking, but product docs say 16k jumbo packets
  are supported by the hardware

netcp: min_mtu 68, max_mtu 9486
- remove netcp_ndo_change_mtu as it is now redundant

via-velocity: min_mtu 64, max_mtu 9000

octeon: min_mtu 46, max_mtu 65370

CC: netdev@vger.kernel.org
CC: Mark Einon <mark.einon@gmail.com>
CC: Vince Bridgers <vbridger@opensource.altera.com>
CC: Rasesh Mody <rasesh.mody@qlogic.com>
CC: Nicolas Ferre <nicolas.ferre@atmel.com>
CC: Santosh Raspatur <santosh@chelsio.com>
CC: Hariprasad S <hariprasad@chelsio.com>
CC:  Christian Benvenuti <benve@cisco.com>
CC: Sujith Sankar <ssujith@cisco.com>
CC: Govindarajulu Varadarajan <_govind@gmx.com>
CC: Neel Patel <neepatel@cisco.com>
CC: Claudiu Manoil <claudiu.manoil@freescale.com>
CC: Yisen Zhuang <yisen.zhuang@huawei.com>
CC: Salil Mehta <salil.mehta@huawei.com>
CC: Hyong-Youb Kim <hykim@myri.com>
CC: Jakub Kicinski <jakub.kicinski@netronome.com>
CC: Olof Johansson <olof@lixom.net>
CC: Jiri Pirko <jiri@resnulli.us>
CC: Byungho An <bh74.an@samsung.com>
CC: Girish K S <ks.giri@samsung.com>
CC: Vipul Pandya <vipul.pandya@samsung.com>
CC: Giuseppe Cavallaro <peppe.cavallaro@st.com>
CC: Alexandre Torgue <alexandre.torgue@st.com>
CC: Andy Gospodarek <andy@greyhouse.net>
CC: Wingman Kwok <w-kwok2@ti.com>
CC: Murali Karicheri <m-karicheri2@ti.com>
CC: Francois Romieu <romieu@fr.zoreil.com>
Signed-off-by: NJarod Wilson <jarod@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add missing ABI defines and eBPF instructions to allow
mark to be passed on and extend prepend parsing on the
RX path to pick it up from packet metadata.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Periodically poll stats and call into offloaded actions
to update them.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add hardware bpf offload on our smart NICs.  Detect if
capable firmware is loaded and use it to load the code JITed
with just added translator onto programmable engines.

This commit only supports offloading cls_bpf in legacy mode
(non-direct action).
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fix to return a negative error code from the error handling
case instead of 0, as done elsewhere in this function.

Fixes: 73725d9d ("nfp: allocate ring SW structs dynamically")
Signed-off-by: NWei Yongjun <weiyongjun1@huawei.com>
Acked-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There is no need to pad frames to ETH_ZLEN on RX.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NSimon Horman <simon.horman@netronome.com>
Reviewed-by: NDinan Gunawardena <dinan.gunawardena@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Remove unnecessary version.h includes.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NSimon Horman <simon.horman@netronome.com>
Reviewed-by: NDinan Gunawardena <dinan.gunawardena@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

idx can be returned as -ENOSPC, so we should check for this first
before using it as an index into nn->vxlan_usecnt[] to avoid an
out of bounds array offset read.
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Acked-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

nfp_net_write_mac_addr() always writes to the BAR the current
device address taken from netdev struct.  The address given
as parameter is actually ignored.  Since all callers pass
netdev->dev_addr simply remove the parameter.

While at it improve the function's kdoc a bit.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This change replaces the network device operations for adding or removing a
VXLAN port with operations that are more generically defined to be used for
any UDP offload port but provide a type.  As such by just adding a line to
verify that the offload type is VXLAN we can maintain the same
functionality.
Signed-off-by: NAlexander Duyck <aduyck@mirantis.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We were using an incorrect define to get the irq vector number.
NFP_NET_CFG_LSC is a control BAR offset, LSC interrupt vector
index is called NFP_NET_IRQ_LSC_IDX.  For machines with less
than 30 CPUs this meant that we were disabling/enabling IRQ 0.
For bigger hosts we were just playing with the 31st RX/TX
interrupt.

Fixes: 0ba40af9 ("nfp: move link state interrupt request/free calls")
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Some callers of nfp_net_reconfig() are in atomic context so
we used to busy wait for commands to complete.  In worst case
scenario that means locking up a core for up to 5 seconds
when a command times out.  Lets add a timer-based mechanism
of asynchronously checking whether reconfiguration completed
successfully for atomic callers to use.  Non-atomic callers
can now just sleep.

The approach taken is quite simple because (1) synchronous
reconfigurations always happen under RTNL (or before device
is registered); (2) we can coalesce pending reconfigs.
There is no need for request queues, timer which eventually
takes a look at reconfiguration result to report errors is
good enough.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Meaning of data_len and meta_len RX WB descriptor fields is
slightly confusing.  Add a comment with a diagram clarifying
the layout.  Also remove the buffer length validation:
(a) it's imprecise for static rx-offsets; (b) if firmware
is buggy enough to DMA past the end of the buffer
WARN_ON_ONCE() doesn't seem like a strong enough response.
skb_put() will do the checking for us anyway.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Since much of the required changes have already been made for
changing MTU at runtime let's use it for ring size changes as
well.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Soon ring resize will call this functions with values
different than the current configuration we need to
explicitly pass the ring count as parameter.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When changing MTU on running device first allocate new rings
and buffers and once it succeeds proceed with changing MTU.

Allocation of new rings is not really necessary for this
operation - it's done to keep the code simple and because
size of the extra ring memory is quite small compared to
the size of buffers.

Operation can still fail midway through if FW communication
times out.  In that case we retry with old MTU (rings).
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Free list buffer size needs to be propagated to few functions
as a parameter and added to struct nfp_net_rx_ring since soon
some of the functions will be reused to manage rings with
buffers of size different than nn->fl_bufsz.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

FW reconfiguration in .ndo_open()/.ndo_stop() should reset/
restore queue state.  Since we need IRQs to be disabled when
filling rings on RX path we have to move disable_irq() from
.ndo_open() all the way up to IRQ allocation.

nfp_net_start_vec() becomes trivial now so it's inlined.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Divide .ndo_open() and .ndo_stop() into logical, callable
chunks.  No functional changes.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

nfp_net_[rt]x_ring_{alloc,free} should only allocate or free
ring resources without touching the device.  Move setting
parameters in the BAR to separate functions.  This will make
it possible to reuse alloc/free functions to allocate new
rings while the device is running.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We want the .ndo_open() to have following structure:
 - allocate resources;
 - configure HW/FW;
 - enable the device from stack perspective.
Therefore filling RX rings needs to be moved to the beginning
of .ndo_open().
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Separate allocation of buffers from giving them to FW,
thanks to this it will be possible to move allocation
earlier on .ndo_open() path and reuse buffers during
runtime reconfiguration.

Similar to TX side clean up the spill of functionality
from flush to freeing the ring.  Unlike on TX side,
RX ring reset does not free buffers from the ring.
Ring reset means only that FW pointers are zeroed and
buffers on the ring must be placed in [0, cnt - 1)
positions.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Since we never used flush without freeing the ring later
the functionality of the two operations is mixed.
Rename flush to ring reset and move there all the things
which have to be done after FW ring state is cleared.
While at it do some clean-ups.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

To be able to switch rings more easily on config changes
allocate them dynamically, separately from nfp_net structure.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

nfp_net_[rt]x_ring_init functions used to be called from probe
path only and some of their functionality was spilled to the
call site.  In order to reuse them for ring reconfiguration
we need them to do all the init.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

nfp_net_{alloc|free}_rings contained strange mix of allocations
and vector initialization.  Remove it, declare vector init as
a separate function and handle allocations explicitly.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We need to be able to disable the link state interrupt when
the device is brought down.  We used to just free the IRQ
at the beginning of .ndo_stop().  As we now move towards
more ordered .ndo_open()/.ndo_stop() paths LSC allocation
should be placed in the "allocate resource" section.

Since the IRQ can't be freed early in .ndo_stop(), it is
disabled instead.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When calculating the RX buffer length we need to account
for up to 2 VLAN tags.  Rounding up to 1k is an relic of
a distant past and can be removed.  While at it also remove
trivial print statement.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Netdevs default to carrier on, we should call netif_carrier_off()
during initialization since we handle carrier state changes in the
driver.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NRolf Neugebauer <rolf.neugebauer@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We need to clear delayed kick counters when we free rings otherwise
after ndo_close()/ndo_open() we could kick HW by more entries than
actually written to rings.
Signed-off-by: NJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: NRolf Neugebauer <rolf.neugebauer@netronome.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The name NETIF_F_ALL_CSUM is a misnomer. This does not correspond to the
set of features for offloading all checksums. This is a mask of the
checksum offload related features bits. It is incorrect to set both
NETIF_F_HW_CSUM and NETIF_F_IP_CSUM or NETIF_F_IPV6 at the same time for
features of a device.

This patch:
  - Changes instances of NETIF_F_ALL_CSUM to NETIF_F_CSUM_MASK (where
    NETIF_F_ALL_CSUM is being used as a mask).
  - Changes bonding, sfc/efx, ipvlan, macvlan, vlan, and team drivers to
    use NEITF_F_HW_CSUM in features list instead of NETIF_F_ALL_CSUM.
Signed-off-by: NTom Herbert <tom@herbertland.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>