Control queue is fairly low latency, and requires SKB allocations,
which means we can't even reach 0.5Msps with perf events. Allow
perf events to be delivered to data queues. This allows us to not
only use multiple queues, but also receive and deliver to user space
more than 5Msps per queue (Xeon E5-2630 v4 2.20GHz, no retpolines).
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
In preparation for SKB-less perf event handling make
nfp_bpf_event_output() take buffer address and length,
not SKB as parameters.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Port id 0xffffffff is reserved for control messages. Allow reception
of messages with this id on data queues. Hand off a raw buffer to
the higher layer code, without allocating SKB for max efficiency.
The RX handle can't modify or keep the buffer, after it returns
buffer is handed back over to the NIC RX free buffer list.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Representor packets are received on PF queues with special metadata tag
for demux. There is no reason to resolve the representor ID -> netdev
after the skb has been allocated. Move the code, this will allow us to
handle special FW messages without SKB allocation overhead.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
After device is stopped we reset the rings by moving all free buffers
to positions [0, cnt - 2], and clear the position cnt - 1 in the ring.
We then proceed to clear the read/write pointers. This means that if
we try to reset the ring again the code will assume that the next to
fill buffer is at position 0 and swap it with cnt - 1. Since we
previously cleared position cnt - 1 it will lead to leaking the first
buffer and leaving ring in a bad state.
This scenario can only happen if FW communication fails, in which case
the ring will never be used again, so the fact it's in a bad state will
not be noticed. Buffer leak is the only problem. Don't try to move
buffers in the ring if the read/write pointers indicate the ring was
never used or have already been reset.
nfp_net_clear_config_and_disable() is now fully idempotent.
Found by code inspection, FW communication failures are very rare,
and reconfiguring a live device is not common either, so it's unlikely
anyone has ever noticed the leak.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now that we have offload replay infrastructure added by
commit 326367427c ("net: sched: call reoffload op on block callback reg")
and flows are guaranteed to be removed correctly, we can revert
commit 951a8ee6de ("nfp: reject binding to shared blocks").
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: John Hurley <john.hurley@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann says:
====================
pull-request: bpf-next 2018-07-20
The following pull-request contains BPF updates for your *net-next* tree.
The main changes are:
1) Add sharing of BPF objects within one ASIC: this allows for reuse of
the same program on multiple ports of a device, and therefore gains
better code store utilization. On top of that, this now also enables
sharing of maps between programs attached to different ports of a
device, from Jakub.
2) Cleanup in libbpf and bpftool's Makefile to reduce unneeded feature
detections and unused variable exports, also from Jakub.
3) First batch of RCU annotation fixes in prog array handling, i.e.
there are several __rcu markers which are not correct as well as
some of the RCU handling, from Roman.
4) Two fixes in BPF sample files related to checking of the prog_cnt
upper limit from sample loader, from Dan.
5) Minor cleanup in sockmap to remove a set but not used variable,
from Colin.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Allow program sharing between netdevs of the same NFP ASIC.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Create a higher-level entity to represent a device/ASIC to allow
programs and maps to be shared between device ports. The extra
work is required to make sure we don't destroy BPF objects as
soon as the netdev for which they were loaded gets destroyed,
as other ports may still be using them. When netdev goes away
all of its BPF objects will be moved to other netdevs of the
device, and only destroyed when last netdev is unregistered.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Currently we have two lists of offloaded objects - programs and maps.
Netdevice unregister notifier scans those lists to orphan objects
associated with device being unregistered. This puts unnecessary
(even if negligible) burden on all netdev unregister calls in BPF-
-enabled kernel. The lists of objects may potentially get long
making the linear scan even more problematic. There haven't been
complaints about this mechanisms so far, but it is suboptimal.
Instead of relying on notifiers, make the few BPF-capable drivers
register explicitly for BPF offloads. The programs and maps will
now be collected per-device not on a global list, and only scanned
for removal when driver unregisters from BPF offloads.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
BPF code should unregister the offload capabilities from .ndo_uninit(),
to make sure the operation is atomic with unlist_netdevice(). Plumb
the init/uninit NDOs for vNICs and representors.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Daniel Borkmann says:
====================
pull-request: bpf-next 2018-07-15
The following pull-request contains BPF updates for your *net-next* tree.
The main changes are:
1) Various different arm32 JIT improvements in order to optimize code emission
and make the JIT code itself more robust, from Russell.
2) Support simultaneous driver and offloaded XDP in order to allow for advanced
use-cases where some work is offloaded to the NIC and some to the host. Also
add ability for bpftool to load programs and maps beyond just the cgroup case,
from Jakub.
3) Add BPF JIT support in nfp for multiplication as well as division. For the
latter in particular, it uses the reciprocal algorithm to emulate it, from Jiong.
4) Add BTF pretty print functionality to bpftool in plain and JSON output
format, from Okash.
5) Add build and installation to the BPF helper man page into bpftool, from Quentin.
6) Add a TCP BPF callback for listening sockets which is triggered right after
the socket transitions to TCP_LISTEN state, from Andrey.
7) Add a new cgroup tree command to bpftool which iterates over the whole cgroup
tree and prints all attached programs, from Roman.
8) Improve xdp_redirect_cpu sample to support parsing of double VLAN tagged
packets, from Jesper.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Split handling of offloaded and driver programs completely. Since
offloaded programs always come with XDP_FLAGS_HW_MODE set in reality
there could be no sharing, anyway, programs would only be installed
in driver or in hardware. Splitting the handling allows us to install
programs in HW and in driver at the same time.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Split the query of HW-attached program from the software one.
Introduce new .ndo_bpf command to query HW-attached program.
This will allow drivers to install different programs in HW
and SW at the same time. Netlink can now also carry multiple
programs on dump (in which case mode will be set to
XDP_ATTACHED_MULTI and user has to check per-attachment point
attributes, IFLA_XDP_PROG_ID will not be present). We reuse
IFLA_XDP_PROG_ID skb space for second mode, so rtnl_xdp_size()
doesn't need to be updated.
Note that the installation side is still not there, since all
drivers currently reject installing more than one program at
the time.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Basic operations drivers perform during xdp setup and query can
be moved to helpers in the core. Encapsulate program and flags
into a structure and add helpers. Note that the structure is
intended as the "main" program information source in the driver.
Most drivers will additionally place the program pointer in their
fast path or ring structures.
The helpers don't have a huge impact now, but they will
decrease the code duplication when programs can be installed
in HW and driver at the same time. Encapsulating the basic
operations in helpers will hopefully also reduce the number
of changes to drivers which adopt them.
Helpers could really be static inline, but they depend on
definition of struct netdev_bpf which means they'd have
to be placed in netdevice.h, an already 4500 line header.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
prog_attached of struct netdev_bpf should have been superseded
by simply setting prog_id long time ago, but we kept it around
to allow offloading drivers to communicate attachment mode (drv
vs hw). Subsequently drivers were also allowed to report back
attachment flags (prog_flags), and since nowadays only programs
attached will XDP_FLAGS_HW_MODE can get offloaded, we can tell
the attachment mode from the flags driver reports. Remove
prog_attached member.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
getnstimeofday64 is deprecated in favor of the ktime_get() family of
functions. The direct replacement would be ktime_get_real_ts64(),
but I'm picking the basic ktime_get() instead:
- using a ktime_t simplifies the code compared to timespec64
- using monotonic time instead of real time avoids issues caused
by a concurrent settimeofday() or during a leap second adjustment.
Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
-----BEGIN PGP SIGNATURE-----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=IOF/
-----END PGP SIGNATURE-----
Merge tag 'pci-v4.18-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci
Pull PCI fixes from Bjorn Helgaas:
- Fix a use-after-free in the endpoint code (Dan Carpenter)
- Stop defaulting CONFIG_PCIE_DW_PLAT_HOST to yes (Geert Uytterhoeven)
- Fix an nfp regression caused by a change in how we limit the number
of VFs we can enable (Jakub Kicinski)
- Fix failure path cleanup issues in the new R-Car gen3 PHY support
(Marek Vasut)
- Fix leaks of OF nodes in faraday, xilinx-nwl, xilinx (Nicholas Mc
Guire)
* tag 'pci-v4.18-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
nfp: stop limiting VFs to 0
PCI/IOV: Reset total_VFs limit after detaching PF driver
PCI: faraday: Add missing of_node_put()
PCI: xilinx-nwl: Add missing of_node_put()
PCI: xilinx: Add missing of_node_put()
PCI: endpoint: Use after free in pci_epf_unregister_driver()
PCI: controller: dwc: Do not let PCIE_DW_PLAT_HOST default to yes
PCI: rcar: Clean up PHY init on failure
PCI: rcar: Shut the PHY down in failpath
As we are doing JIT, we would want to use the advanced version of the
reciprocal divide (reciprocal_value_adv) to trade performance with host.
We could reduce the required ALU instructions from 4 to 2 or 1.
Signed-off-by: Jiong Wang <jiong.wang@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
NFP doesn't have integer divide instruction, this patch use reciprocal
algorithm (the basic one, reciprocal_div) to emulate it.
For each u32 divide, we would need 11 instructions to finish the operation.
7 (for multiplication) + 4 (various ALUs) = 11
Given NFP only supports multiplication no bigger than u32, we'd require
divisor and dividend no bigger than that as well.
Also eBPF doesn't support signed divide and has enforced this on C language
level by failing compilation. However LLVM assembler hasn't enforced this,
so it is possible for negative constant to leak in as a BPF_K operand
through assembly code, we reject such cases as well.
Signed-off-by: Jiong Wang <jiong.wang@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
NFP supports u16 and u32 multiplication. Multiplication is done 8-bits per
step, therefore we need 2 steps for u16 and 4 steps for u32.
We also need one start instruction to initialize the sequence and one or
two instructions to fetch the result depending on either you need the high
halve of u32 multiplication.
For ALU64, if either operand is beyond u32's value range, we reject it. One
thing to note, if the source operand is BPF_K, then we need to check "imm"
field directly, and we'd reject it if it is negative. Because for ALU64,
"imm" (with s32 type) is expected to be sign extended to s64 which NFP mul
doesn't support. For ALU32, it is fine for "imm" be negative though,
because the result is 32-bits and here is no difference on the low halve
of result for signed/unsigned mul, so we will get correct result.
Signed-off-by: Jiong Wang <jiong.wang@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
NFP verifier hook is coping range information of the shift amount for
indirect shift operation so optimized shift sequences could be generated.
We want to use range info to do more things. For example, to decide whether
multiplication and divide are supported on the given range.
This patch simply let NFP verifier hook to copy range info for all operands
of all ALU operands.
Signed-off-by: Jiong Wang <jiong.wang@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
The two fields are a copy of umin and umax info of bpf_insn->src_reg
generated by verifier.
Rename to make their meaning clear.
Signed-off-by: Jiong Wang <jiong.wang@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Daniel Borkmann says:
====================
pull-request: bpf-next 2018-07-03
The following pull-request contains BPF updates for your *net-next* tree.
The main changes are:
1) Various improvements to bpftool and libbpf, that is, bpftool build
speed improvements, missing BPF program types added for detection
by section name, ability to load programs from '.text' section is
made to work again, and better bash completion handling, from Jakub.
2) Improvements to nfp JIT's map read handling which allows for optimizing
memcpy from map to packet, from Jiong.
3) New BPF sample is added which demonstrates XDP in combination with
bpf_perf_event_output() helper to sample packets on all CPUs, from Toke.
4) Add a new BPF kselftest case for tracking connect(2) BPF hooks
infrastructure in combination with TFO, from Andrey.
5) Extend the XDP/BPF xdp_rxq_info sample code with a cmdline option to
read payload from packet data in order to use it for benchmarking.
Also for '--action XDP_TX' option implement swapping of MAC addresses
to avoid drops on some hardware seen during testing, from Jesper.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Simple overlapping changes in stmmac driver.
Adjust skb_gro_flush_final_remcsum function signature to make GRO list
changes in net-next, as per Stephen Rothwell's example merge
resolution.
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann says:
====================
pull-request: bpf 2018-07-01
The following pull-request contains BPF updates for your *net* tree.
The main changes are:
1) A bpf_fib_lookup() helper fix to change the API before freeze to
return an encoding of the FIB lookup result and return the nexthop
device index in the params struct (instead of device index as return
code that we had before), from David.
2) Various BPF JIT fixes to address syzkaller fallout, that is, do not
reject progs when set_memory_*() fails since it could still be RO.
Also arm32 JIT was not using bpf_jit_binary_lock_ro() API which was
an issue, and a memory leak in s390 JIT found during review, from
Daniel.
3) Multiple fixes for sockmap/hash to address most of the syzkaller
triggered bugs. Usage with IPv6 was crashing, a GPF in bpf_tcp_close(),
a missing sock_map_release() routine to hook up to callbacks, and a
fix for an omitted bucket lock in sock_close(), from John.
4) Two bpftool fixes to remove duplicated error message on program load,
and another one to close the libbpf object after program load. One
additional fix for nfp driver's BPF offload to avoid stopping offload
completely if replace of program failed, from Jakub.
5) Couple of BPF selftest fixes that bail out in some of the test
scripts if the user does not have the right privileges, from Jeffrin.
6) Fixes in test_bpf for s390 when CONFIG_BPF_JIT_ALWAYS_ON is set
where we need to set the flag that some of the test cases are expected
to fail, from Kleber.
7) Fix to detangle BPF_LIRC_MODE2 dependency from CONFIG_CGROUP_BPF
since it has no relation to it and lirc2 users often have configs
without cgroups enabled and thus would not be able to use it, from Sean.
8) Fix a selftest failure in sockmap by removing a useless setrlimit()
call that would set a too low limit where at the same time we are
already including bpf_rlimit.h that does the job, from Yonghong.
9) Fix BPF selftest config with missing missing NET_SCHED, from Anders.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently the NFP fw only supports L3/L4 hashing so rejects the offload of
filters that output to LAG ports implementing other hash algorithms. Team,
however, uses a BPF function for the hash that is not defined. To support
Team offload, accept hashes that are defined as 'unknown' (only Team
defines such hash types). In this case, use the NFP default of L3/L4
hashing for egress port selection.
Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Extract the tos and the tunnel flags from the tunnel key and offload these
action fields. Only the checksum and tunnel key flags are implemented in
fw so reject offloads of other flags. The tunnel key flag is always
considered set in the fw so enforce that it is set in the rule. Note that
the compulsory setting of the tunnel key flag and optional setting of
checksum is inline with how tc currently generates ipv4 udp tunnel
actions.
Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Previously the ttl for ipv4 udp tunnels was set to the namespace default.
Modify this to attempt to extract the ttl from a full route lookup on the
tunnel destination. If this is not possible then resort to the default.
Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Hardware will automatically update csum in headers when a set action has
been performed. This means we could in the driver ignore the explicit
checksum action when performing a set action.
Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We used to leave bus-info in ethtool driver info empty for
representors in case multi-PCIe-to-single-host cards make
the association between PCIe device and NFP many to one.
It seems these attempts are futile, we need to link the
representors to one PCIe device in sysfs to get consistent
naming, plus devlink uses one PCIe as a handle, anyway.
The multi-PCIe-to-single-system support won't be clean,
if it ever comes.
Turns out some user space (RHEL tests) likes to read bus-info
so just populate it.
While at it remove unnecessary app NULL-check, representors
are spawned by an app, so it must exist.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use napi_consume_skb() in nfp_net_tx_complete() to get bulk free.
Pass 0 as budget for ctrl queue completion since it runs out of
a tasklet.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
NFP NAPI handling will only complete the TXed packets when called
with budget of 0, implement ndo_poll_controller by scheduling NAPI
on all TX queues.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
On some platforms with broken ACPI tables we may not have access
to the Serial Number PCIe capability. This capability is crucial
for us for switchdev operation as we use serial number as switch ID,
and for communication with management FW where interface ID is used.
If we can't determine the Serial Number we have to fail device probe.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After user changes the ring count statistics for deactivated
rings disappear from ethtool -S output. This causes loss of
information to the user and means that ethtool stats may not
add up to interface stats. Always expose counters from all
the rings. Note that we allocate at most num_possible_cpus()
rings so number of rings should be reasonable.
The alternative of only listing stats for rings which were
ever in use could be confusing.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Before 8d85a7a4f2 ("PCI/IOV: Allow PF drivers to limit total_VFs to 0"),
pci_sriov_set_totalvfs(pdev, 0) meant "we can enable TotalVFs virtual
functions". After 8d85a7a4f2, it means "we can't enable *any* VFs".
That broke this scenario where nfp intends to remove any limit on the
number of VFs that can be enabled:
nfp_pci_probe
nfp_pcie_sriov_read_nfd_limit
nfp_rtsym_read_le("nfd_vf_cfg_max_vfs", &err)
pci_sriov_set_totalvfs(pf->pdev, 0) # if FW didn't expose a limit
...
# userspace writes N to sysfs "sriov_numvfs":
sriov_numvfs_store
pci_sriov_get_totalvfs # now returns 0
return -ERANGE
Prior to 8d85a7a4f2, pci_sriov_get_totalvfs() returned TotalVFs, but it
now returns 0.
Remove the pci_sriov_set_totalvfs(pdev, 0) calls so we don't limit the
number of VFs that can be enabled.
Fixes: 8d85a7a4f2 ("PCI/IOV: Allow PF drivers to limit total_VFs to 0")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
[bhelgaas: changelog]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Map read has been supported on NFP, this patch enables optimization
for memcpy from map to packet.
This patch also fixed one latent bug which will cause copying from
unexpected address once memcpy for map pointer enabled. The fixed
code path was not exercised before.
Reported-by: Mary Pham <mary.pham@netronome.com>
Reported-by: David Beckett <david.beckett@netronome.com>
Signed-off-by: Jiong Wang <jiong.wang@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
sizeof() will return unsigned value so in the error check
negative error code will be always larger than sizeof().
Fixes: a0d8e02c35 ("nfp: add support for reading nffw info")
Signed-off-by: Chengguang Xu <cgxu519@gmx.com>
Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
TC shared blocks allow multiple qdiscs to be grouped together and filters
shared between them. Currently the chains of filters attached to a block
are only flushed when the block is removed. If a qdisc is removed from a
block but the block still exists, flow del messages are not passed to the
callback registered for that qdisc. For the NFP, this presents the
possibility of rules still existing in hw when they should be removed.
Prevent binding to shared blocks until the kernel can send per qdisc del
messages when block unbinds occur.
tcf_block_shared() was not used outside of the core until now, so also
add an empty implementation for builds with CONFIG_NET_CLS=n.
Fixes: 4861738775 ("net: sched: introduce shared filter blocks infrastructure")
Signed-off-by: John Hurley <john.hurley@netronome.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Previously it was not possible to distinguish between mpls ether types and
other ether types. This leads to incorrect classification of offloaded
filters that match on mpls ether type. For example the following two
filters overlap:
# tc filter add dev eth0 parent ffff: \
protocol 0x8847 flower \
action mirred egress redirect dev eth1
# tc filter add dev eth0 parent ffff: \
protocol 0x0800 flower \
action mirred egress redirect dev eth2
The driver now correctly includes the mac_mpls layer where HW stores mpls
fields, when it detects an mpls ether type. It also sets the MPLS_Q bit to
indicate that the filter should match mpls packets.
Fixes: bb055c198d ("nfp: add mpls match offloading support")
Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pass the extact struct from a tc qdisc add to the block bind function and,
in turn, to the setup_tc ndo of binding device via the tc_block_offload
struct. Pass this back to any block callback registrations to allow
netlink logging of fails in the bind process.
Signed-off-by: John Hurley <john.hurley@netronome.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stopping offload completely if replace of program failed dates
back to days of transparent offload. Back then we wanted to
silently fall back to the in-driver processing. Today we mark
programs for offload when they are loaded into the kernel, so
the transparent offload is no longer a reality.
Flags check in the driver will only allow replace of a driver
program with another driver program or an offload program with
another offload program.
When driver program is replaced stopping offload is a no-op,
because driver program isn't offloaded. When replacing
offloaded program if the offload fails the entire operation
will fail all the way back to user space and we should continue
using the old program. IOW when replacing a driver program
stopping offload is unnecessary and when replacing offloaded
program - it's a bug, old program should continue to run.
In practice this bug would mean that if offload operation was to
fail (either due to FW communication error, kernel OOM or new
program being offloaded but for a different netdev) driver
would continue reporting that previous XDP program is offloaded
but in fact no program will be loaded in hardware. The failure
is fairly unlikely (found by inspection, when working on the code)
but it's unpleasant.
Backport note: even though the bug was introduced in commit
cafa92ac25 ("nfp: bpf: add support for XDP_FLAGS_HW_MODE"),
this fix depends on commit 441a33031f ("net: xdp: don't allow
device-bound programs in driver mode"), so this fix is sufficient
only in v4.15 or newer. Kernels v4.13.x and v4.14.x do need to
stop offload if it was transparent/opportunistic, i.e. if
XDP_FLAGS_HW_MODE was not set on running program.
Fixes: cafa92ac25 ("nfp: bpf: add support for XDP_FLAGS_HW_MODE")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Currently the default case is not handled, which with future command
introductions would introduce a warning. So handle it.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull networking fixes from David Miller:
1) Various netfilter fixlets from Pablo and the netfilter team.
2) Fix regression in IPVS caused by lack of PMTU exceptions on local
routes in ipv6, from Julian Anastasov.
3) Check pskb_trim_rcsum for failure in DSA, from Zhouyang Jia.
4) Don't crash on poll in TLS, from Daniel Borkmann.
5) Revert SO_REUSE{ADDR,PORT} change, it regresses various things
including Avahi mDNS. From Bart Van Assche.
6) Missing of_node_put in qcom/emac driver, from Yue Haibing.
7) We lack checking of the TCP checking in one special case during SYN
receive, from Frank van der Linden.
8) Fix module init error paths of mac80211 hwsim, from Johannes Berg.
9) Handle 802.1ad properly in stmmac driver, from Elad Nachman.
10) Must grab HW caps before doing quirk checks in stmmac driver, from
Jose Abreu.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (81 commits)
net: stmmac: Run HWIF Quirks after getting HW caps
neighbour: skip NTF_EXT_LEARNED entries during forced gc
net: cxgb3: add error handling for sysfs_create_group
tls: fix waitall behavior in tls_sw_recvmsg
tls: fix use-after-free in tls_push_record
l2tp: filter out non-PPP sessions in pppol2tp_tunnel_ioctl()
l2tp: reject creation of non-PPP sessions on L2TPv2 tunnels
mlxsw: spectrum_switchdev: Fix port_vlan refcounting
mlxsw: spectrum_router: Align with new route replace logic
mlxsw: spectrum_router: Allow appending to dev-only routes
ipv6: Only emit append events for appended routes
stmmac: added support for 802.1ad vlan stripping
cfg80211: fix rcu in cfg80211_unregister_wdev
mac80211: Move up init of TXQs
mac80211_hwsim: fix module init error paths
cfg80211: initialize sinfo in cfg80211_get_station
nl80211: fix some kernel doc tag mistakes
hv_netvsc: Fix the variable sizes in ipsecv2 and rsc offload
rds: avoid unenecessary cong_update in loop transport
l2tp: clean up stale tunnel or session in pppol2tp_connect's error path
...
We need to release the refcnt on dst_entry in the route table, otherwise
we will leak the route.
Fixes: 8e6a9046b6 ("nfp: flower vxlan neighbour offload")
Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com>
Signed-off-by: Louis Peens <louis.peens@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
.ndo_get_phys_port_name was recently extended to support multi-vNIC
FWs. These are firmwares which can have more than one vNIC per PF
without associated port (e.g. Adaptive Buffer Management FW), therefore
we need a way of distinguishing the vNICs. Unfortunately, it's too
late to make flower use the same naming. Flower users may depend on
.ndo_get_phys_port_name returning -EOPNOTSUPP, for example the name
udev gave the PF vNIC was just the bare PCI device-based name before
the change, and will have 'nn0' appended after.
To ensure flower's vNIC doesn't have phys_port_name attribute, add
a flag to vNIC struct and set it in flower code. New projects will
not set the flag adhere to the naming scheme from the start.
Fixes: 51c1df83e3 ("nfp: assign vNIC id as phys_port_name of vNICs which are not ports")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We are gathering software statistics on per-ring basis.
.ndo_get_stats64 handler adds the rings up. Unfortunately
we are currently only adding up active rings, which means
that if user decreases the number of active rings the
statistics from deactivated rings will no longer be counted
and total interface statistics may go backwards.
Always sum all possible rings, the stats are allocated
statically for max number of rings, so we don't have to
worry about them being removed. We could add the stats
up when user changes the ring count, but it seems unnecessary..
Adding up inactive rings will be very quick since no datapath
will be touching them.
Fixes: 164d1e9e5d ("nfp: add support for ethtool .set_channels")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
