Add a test which checks that the RSS table is at least 4x the max
queue count supported by the device. The original RSS spec from
Microsoft stated that the RSS indirection table should be 2 to 8
times the CPU count, presumably assuming queue per CPU. If the
CPU count is not a power of two, however, a power-of-2 table
2x larger than queue count results in a 33% traffic imbalance.
Validate that the indirection table is at least 4x the queue
count. This lowers the imbalance to 16% which empirically
appears to be more acceptable to memcache-like workloads.
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20260131225454.1225151-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This reverts commit 77b9c4a438, reversing
changes made to 4515ec4ad5:
931420a2fc ("selftests/net: Add netkit container tests")
ab771c938d ("selftests/net: Make NetDrvContEnv support queue leasing")
6be87fbb27 ("selftests/net: Add env for container based tests")
61d99ce3df ("selftests/net: Add bpf skb forwarding program")
920da36341 ("netkit: Add xsk support for af_xdp applications")
eef51113f8 ("netkit: Add netkit notifier to check for unregistering devices")
b5ef109d22 ("netkit: Implement rtnl_link_ops->alloc and ndo_queue_create")
b5c3fa4a0b ("netkit: Add single device mode for netkit")
0073d2fd67 ("xsk: Proxy pool management for leased queues")
1ecea95dd3 ("xsk: Extend xsk_rcv_check validation")
804bf334d0 ("net: Proxy netdev_queue_get_dma_dev for leased queues")
0caa9a8dde ("net: Proxy net_mp_{open,close}_rxq for leased queues")
ff8889ff91 ("net, ethtool: Disallow leased real rxqs to be resized")
9e2103f361 ("net: Add lease info to queue-get response")
31127dedde ("net: Implement netdev_nl_queue_create_doit")
a5546e18f7 ("net: Add queue-create operation")
The series will conflict with io_uring work, and the code needs more
polish.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add two tests using NetDrvContEnv. One basic test that sets up a netkit
pair, with one end in a netns. Use LOCAL_PREFIX_V6 and nk_forward BPF
program to ping from a remote host to the netkit in netns.
Second is a selftest for netkit queue leasing, using io_uring zero copy
test binary inside of a netns with netkit. This checks that memory
providers can be bound against virtual queues in a netkit within a
netns that are leasing from a physical netdev in the default netns.
Signed-off-by: David Wei <dw@davidwei.uk>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Link: https://patch.msgid.link/20260115082603.219152-17-daniel@iogearbox.net
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Looks like the liburing is not updated by distros very aggressively.
Presumably because a lot of packages depend on it. I just updated
to Fedora 43 and it's still on liburing 2.9. The test is 9mo old,
at this stage I think this warrants handling the build failure
more gracefully.
Detect if iouring is recent enough and if not print a warning
and exclude the C prog from build. The Python test will just
fail since the binary won't exist. But it removes the major
annoyance of having to update liburing from sources when
developing other tests.
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20251121040259.3647749-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Rewrite the existing toeplitz.sh test in Python. The conversion
is a lot less exact than the GRO one. We use Netlink APIs to
get the device RSS and IRQ information. We expect that the device
has neither RPS nor RFS configured, and set RPS up as part of
the test.
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20251120021024.2944527-11-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The GRO test can run on a real device or a veth.
The Toeplitz hash test can only run on a real device.
Move them from net/ to drivers/net/ and drivers/net/hw/ respectively.
There are two scripts which set up the environment for these tests
setup_loopback.sh and setup_veth.sh. Move those scripts to net/lib.
The paths to the setup files are a little ugly but they will be
deleted shortly.
toeplitz_client.sh is not a test in itself, but rather a helper
to send traffic, so add it to TEST_FILES rather than TEST_PROGS.
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20251120021024.2944527-6-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
We get a significant number of conflicts between net and net-next
because of selftests Makefile changes. People tend to append new
test cases at the end of the Makefile when there's no clear sort
order. Sort all networking selftests Makefiles, use the following
format:
VAR_NAME := \
entry1 \
entry2 \
entry3 \
# end of VAR_NAME
Some Makefiles are already pretty close to this.
Acked-by: Antonio Quartulli <antonio@openvpn.net>
Acked-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Acked-by: Allison Henderson <allison.henderson@oracle.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Link: https://patch.msgid.link/20251003210127.1021918-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add simple tests to validate that the driver sets up timestamping
configuration according to what is reported in capabilities.
For RX timestamping we allow driver to fallback to wider scope for
timestamping if filter is applied. That actually means that driver
can enable ptpv2-event when it reports ptpv2-l4-event is supported,
but not vice versa.
Signed-off-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>
Link: https://patch.msgid.link/20250923173310.139623-5-vadim.fedorenko@linux.dev
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add a simple test for checking that RSS on flow label works,
and that its rejected for IPv4 flows.
# ./tools/testing/selftests/drivers/net/hw/rss_flow_label.py
TAP version 13
1..2
ok 1 rss_flow_label.test_rss_flow_label
ok 2 rss_flow_label.test_rss_flow_label_6only
# Totals: pass:2 fail:0 xfail:0 xpass:0 skip:0 error:0
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Joe Damato <joe@dama.to>
Link: https://patch.msgid.link/20250811234212.580748-5-kuba@kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Revert fbbf93556f ("selftests: nic_performance: Add selftest for performance of NIC driver")
Revert c087dc5439 ("selftests: nic_link_layer: Add selftest case for speed and duplex states")
Revert 6116075e18 ("selftests: nic_link_layer: Add link layer selftest for NIC driver")
These tests don't clean up after themselves, don't use the disruptive
annotations, don't get included in make install etc. etc. The tests
were added before we have any "HW" runner, so the issues were missed.
Our CI doesn't have any way of excluding broken tests, remove these
for now to stop the random pollution of results due to broken env.
We can always add them back once / if fixed.
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Reviewed-by: David Wei <dw@davidwei.uk>
Link: https://patch.msgid.link/20250507140109.929801-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Pull io_uring zero-copy receive support from Jens Axboe:
"This adds support for zero-copy receive with io_uring, enabling fast
bulk receive of data directly into application memory, rather than
needing to copy the data out of kernel memory.
While this version only supports host memory as that was the initial
target, other memory types are planned as well, with notably GPU
memory coming next.
This work depends on some networking components which were queued up
on the networking side, but have now landed in your tree.
This is the work of Pavel Begunkov and David Wei. From the v14 posting:
'We configure a page pool that a driver uses to fill a hw rx queue
to hand out user pages instead of kernel pages. Any data that ends
up hitting this hw rx queue will thus be dma'd into userspace
memory directly, without needing to be bounced through kernel
memory. 'Reading' data out of a socket instead becomes a
_notification_ mechanism, where the kernel tells userspace where
the data is. The overall approach is similar to the devmem TCP
proposal
This relies on hw header/data split, flow steering and RSS to
ensure packet headers remain in kernel memory and only desired
flows hit a hw rx queue configured for zero copy. Configuring this
is outside of the scope of this patchset.
We share netdev core infra with devmem TCP. The main difference is
that io_uring is used for the uAPI and the lifetime of all objects
are bound to an io_uring instance. Data is 'read' using a new
io_uring request type. When done, data is returned via a new shared
refill queue. A zero copy page pool refills a hw rx queue from this
refill queue directly. Of course, the lifetime of these data
buffers are managed by io_uring rather than the networking stack,
with different refcounting rules.
This patchset is the first step adding basic zero copy support. We
will extend this iteratively with new features e.g. dynamically
allocated zero copy areas, THP support, dmabuf support, improved
copy fallback, general optimisations and more'
In a local setup, I was able to saturate a 200G link with a single CPU
core, and at netdev conf 0x19 earlier this month, Jamal reported
188Gbit of bandwidth using a single core (no HT, including soft-irq).
Safe to say the efficiency is there, as bigger links would be needed
to find the per-core limit, and it's considerably more efficient and
faster than the existing devmem solution"
* tag 'for-6.15/io_uring-rx-zc-20250325' of git://git.kernel.dk/linux:
io_uring/zcrx: add selftest case for recvzc with read limit
io_uring/zcrx: add a read limit to recvzc requests
io_uring: add missing IORING_MAP_OFF_ZCRX_REGION in io_uring_mmap
io_uring: Rename KConfig to Kconfig
io_uring/zcrx: fix leaks on failed registration
io_uring/zcrx: recheck ifq on shutdown
io_uring/zcrx: add selftest
net: add documentation for io_uring zcrx
io_uring/zcrx: add copy fallback
io_uring/zcrx: throttle receive requests
io_uring/zcrx: set pp memory provider for an rx queue
io_uring/zcrx: add io_recvzc request
io_uring/zcrx: dma-map area for the device
io_uring/zcrx: implement zerocopy receive pp memory provider
io_uring/zcrx: grab a net device
io_uring/zcrx: add io_zcrx_area
io_uring/zcrx: add interface queue and refill queue
Add tests to check that the napi retained the IRQ after down/up,
multiple changes in the number of rx queues and after
attaching/releasing XDP program.
Tested on ice and idpf:
# NETIF=<iface> tools/testing/selftests/drivers/net/hw/irq.py
KTAP version 1
1..4
ok 1 irq.check_irqs_reported
ok 2 irq.check_reconfig_queues
ok 3 irq.check_reconfig_xdp
ok 4 irq.check_down
# Totals: pass:4 fail:0 xfail:0 xpass:0 skip:0 error:0
Tested-by: Ahmed Zaki <ahmed.zaki@intel.com>
Link: https://patch.msgid.link/20250224232228.990783-7-ahmed.zaki@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add a simple test for TSO. Send a few MB of data and check device
stats to verify that the device was performing segmentation.
Do the same thing over a few tunnel types.
Injecting GSO packets directly would give us more ability to test
corner cases, but perhaps starting simple is good enough?
# ./ksft-net-drv/drivers/net/hw/tso.py
# Detected qstat for LSO wire-packets
KTAP version 1
1..14
ok 1 tso.ipv4 # SKIP Test requires IPv4 connectivity
ok 2 tso.vxlan4_ipv4 # SKIP Test requires IPv4 connectivity
ok 3 tso.vxlan6_ipv4 # SKIP Test requires IPv4 connectivity
ok 4 tso.vxlan_csum4_ipv4 # SKIP Test requires IPv4 connectivity
ok 5 tso.vxlan_csum6_ipv4 # SKIP Test requires IPv4 connectivity
ok 6 tso.gre4_ipv4 # SKIP Test requires IPv4 connectivity
ok 7 tso.gre6_ipv4 # SKIP Test requires IPv4 connectivity
ok 8 tso.ipv6
ok 9 tso.vxlan4_ipv6
ok 10 tso.vxlan6_ipv6
ok 11 tso.vxlan_csum4_ipv6
ok 12 tso.vxlan_csum6_ipv6
# Testing with mangleid enabled
ok 13 tso.gre4_ipv6
ok 14 tso.gre6_ipv6
# Totals: pass:7 fail:0 xfail:0 xpass:0 skip:7 error:0
Note that the test currently depends on the driver reporting
the LSO count via qstat, which appears to be relatively rare
(virtio, cisco/enic, sfc/efc; but virtio needs host support).
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20250218225426.77726-5-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add a selftest for io_uring zero copy Rx. This test cannot run locally
and requires a remote host to be configured in net.config. The remote
host must have hardware support for zero copy Rx as listed in the
documentation page. The test will restore the NIC config back to before
the test and is idempotent.
liburing is required to compile the test and be installed on the remote
host running the test.
Signed-off-by: David Wei <dw@davidwei.uk>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/20250215000947.789731-12-dw@davidwei.uk
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Add selftest case to check the send and receive throughput.
Supported link modes between local NIC driver and partner
are varied. Then send and receive throughput is captured
and verified. Test uses iperf3 tool.
Add iperf3 server/client function in GenerateTraffic class.
Signed-off-by: Mohan Prasad J <mohan.prasad@microchip.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Add selftest file for the link layer tests of a NIC driver.
Test for auto-negotiation is added.
Add LinkConfig class for changing link layer configs.
Selftest makes use of ksft modules and ethtool.
Include selftest file in the Makefile.
Signed-off-by: Mohan Prasad J <mohan.prasad@microchip.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Only RX side for now and small message to test the setup.
In the future, we can extend it to TX side and to testing
both sides with a couple of megs of data.
make \
-C tools/testing/selftests \
TARGETS="drivers/hw/net" \
install INSTALL_PATH=~/tmp/ksft
scp ~/tmp/ksft ${HOST}:
scp ~/tmp/ksft ${PEER}:
cfg+="NETIF=${DEV}\n"
cfg+="LOCAL_V6=${HOST_IP}\n"
cfg+="REMOTE_V6=${PEER_IP}\n"
cfg+="REMOTE_TYPE=ssh\n"
cfg+="REMOTE_ARGS=root@${PEER}\n"
echo -e "$cfg" | ssh root@${HOST} "cat > ksft/drivers/net/net.config"
ssh root@${HOST} "cd ksft && ./run_kselftest.sh -t drivers/net:devmem.py"
Reviewed-by: Mina Almasry <almasrymina@google.com>
Signed-off-by: Stanislav Fomichev <sdf@fomichev.me>
Reviewed-by: Joe Damato <jdamato@fastly.com>
Link: https://patch.msgid.link/20241107181211.3934153-13-sdf@fomichev.me
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add tests focusing on indirection table configuration and
creating extra RSS contexts in drivers which support it.
$ export NETIF=eth0 REMOTE_...
$ ./drivers/net/hw/rss_ctx.py
KTAP version 1
1..8
ok 1 rss_ctx.test_rss_key_indir
ok 2 rss_ctx.test_rss_context
ok 3 rss_ctx.test_rss_context4
# Increasing queue count 44 -> 66
# Failed to create context 32, trying to test what we got
ok 4 rss_ctx.test_rss_context32 # SKIP Tested only 31 contexts, wanted 32
ok 5 rss_ctx.test_rss_context_overlap
ok 6 rss_ctx.test_rss_context_overlap2
# .. sprays traffic like a headless chicken ..
not ok 7 rss_ctx.test_rss_context_out_of_order
ok 8 rss_ctx.test_rss_context4_create_with_cfg
# Totals: pass:6 fail:1 xfail:0 xpass:0 skip:1 error:0
Note that rss_ctx.test_rss_context_out_of_order fails with the device
I tested with, but it seems to be a device / driver bug.
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240626012456.2326192-5-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Run tools/testing/selftest/net/csum.c as part of drv-net.
This binary covers multiple scenarios, based on arguments given,
for both IPv4 and IPv6:
- Accept UDP correct checksum
- Detect UDP invalid checksum
- Accept TCP correct checksum
- Detect TCP invalid checksum
- Transmit UDP: basic checksum offload
- Transmit UDP: zero checksum conversion
The test direction is reversed between receive and transmit tests, so
that the NIC under test is always the local machine.
In total this adds up to 12 testcases, with more to follow. For
conciseness, I replaced individual functions with a function factory.
Also detect hardware offload feature availability using Ethtool
netlink and skip tests when either feature is off. This need may be
common for offload feature tests and eventually deserving of a thin
wrapper in lib.py.
Missing are the PF_PACKET based send tests ('-P'). These use
virtio_net_hdr to program hardware checksum offload. Which requires
looking up the local MAC address and (harder) the MAC of the next hop.
I'll have to give it some though how to do that robustly and where
that code would belong.
Tested:
make -C tools/testing/selftests/ \
TARGETS="drivers/net drivers/net/hw" \
install INSTALL_PATH=/tmp/ksft
cd /tmp/ksft
sudo NETIF=ens4 REMOTE_TYPE=ssh \
REMOTE_ARGS="root@10.40.0.2" \
LOCAL_V4="10.40.0.1" \
REMOTE_V4="10.40.0.2" \
./run_kselftest.sh -t drivers/net/hw:csum.py
Signed-off-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/20240507154216.501111-1-willemdebruijn.kernel@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Bugs in memory allocation failure paths are quite common.
Add a test exercising those paths based on qstat and page pool
failure hook.
Running on bnxt:
# ./drivers/net/hw/pp_alloc_fail.py
KTAP version 1
1..1
# ethtool -G change retval: success
ok 1 pp_alloc_fail.test_pp_alloc
# Totals: pass:1 fail:0 xfail:0 xpass:0 skip:0 error:0
I initially wrote this test to validate commit be43b7489a ("net/mlx5e:
RX, Fix page_pool allocation failure recovery for striding rq") but mlx5
still doesn't have qstat. So I run it on bnxt, and while bnxt survives
I found the problem fixed in commit 7301177307 ("eth: bnxt: fix counting
packets discarded due to OOM and netpoll").
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/20240429144426.743476-7-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
We created a separate directory for HW-only tests, recently.
Glue in the Python test library there, Python is a bit annoying
when it comes to using library code located "lower"
in the directory structure.
Reuse the Env class, but let tests require non-nsim setup.
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/20240429144426.743476-3-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The tests in net/forwarding are generally expected to be HW-independent.
There are however several tests that, while not depending on any HW in
particular, nevertheless depend on being used on HW interfaces. Placing
these selftests to net/forwarding is confusing, because the selftest will
just report it can't be run on veth pairs. At the same time, placing them
to a particular driver's selftests subdirectory would be wrong.
Instead, add a new directory, drivers/net/hw, where these generic but HW
independent selftests should be placed. Move over several such tests
including one helper library.
Since typically these tests will not be expected to run, omit the directory
drivers/net/hw from the TARGETS list in selftests/Makefile. Retain a
Makefile in the new directory itself, so that a user can make -C into that
directory and act on those tests explicitly.
Cc: Roger Quadros <rogerq@kernel.org>
Cc: Tobias Waldekranz <tobias@waldekranz.com>
Cc: Danielle Ratson <danieller@nvidia.com>
Cc: Davide Caratti <dcaratti@redhat.com>
Cc: Johannes Nixdorf <jnixdorf-oss@avm.de>
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Link: https://lore.kernel.org/r/e11dae1f62703059e9fc2240004288ac7cc15756.1711464583.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
