linux/tools/testing/selftests/drivers/net/hw, branch master

selftests: drv-net: rss_ctx: test RSS contexts persist after ifdown/up

2026-02-21T00:34:47Z

Add a test to verify that RSS contexts persist across interface down/up along with their associated Ntuple filters. Another test that creates contexts/rules keeping interface down and test their persistence is also added. Tested on bnxt_en: TAP version 13 1..1 # timeout set to 0 # selftests: drivers/net/hw: rss_ctx.py # TAP version 13 # 1..2 # ok 1 rss_ctx.test_rss_context_persist_create_and_ifdown # ok 2 rss_ctx.test_rss_context_persist_ifdown_and_create # SKIP Create context not supported with interface down # # Totals: pass:1 fail:0 xfail:0 xpass:0 skip:1 error:0 Reviewed-by: Andy Gospodarek Signed-off-by: Pavan Chebbi Signed-off-by: Michael Chan Link: https://patch.msgid.link/20260219185313.2682148-4-michael.chan@broadcom.com Signed-off-by: Jakub Kicinski

selftests: drv-net: add HDS payload sweep test for devmem TCP

2026-02-17T09:40:14Z

Add check_rx_hds test that verifies header/data split works across payload sizes. The test sweeps payload sizes from 1 byte to 8KB, if any data propagates up to userspace as SCM_DEVMEM_LINEAR, then the test fails. This shows that regardless of payload size, ncdevmem's configuration of hds-thresh to 0 is respected. Add -L (--fail-on-linear) flag to ncdevmem that causes the receiver to fail if any SCM_DEVMEM_LINEAR cmsg is received. Use socat option for fixed block sizing and tcp nodelay to disable nagle's algo to avoid buffering. Signed-off-by: Bobby Eshleman Link: https://patch.msgid.link/20260211-fbnic-tcp-hds-fixes-v1-4-55d050e6f606@meta.com Signed-off-by: Paolo Abeni

selftests: drv-net: limit RPS test CPUs to supported range

2026-02-13T02:41:49Z

The _get_unused_cpus() function can return CPU numbers >= 16, which exceeds RPS_MAX_CPUS in toeplitz.c. When this happens, the test fails with a cryptic message: # Exception| Traceback (most recent call last): # Exception| File "/tmp/cur/linux/tools/testing/selftests/net/lib/py/ksft.py", line 319, in ksft_run # Exception| func(*args) # Exception| File "/tmp/cur/linux/tools/testing/selftests/drivers/net/hw/toeplitz.py", line 189, in test # Exception| with bkg(" ".join(rx_cmd), ksft_ready=True, exit_wait=True) as rx_proc: # Exception| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ # Exception| File "/tmp/cur/linux/tools/testing/selftests/net/lib/py/utils.py", line 124, in __init__ # Exception| super().__init__(comm, background=True, # Exception| File "/tmp/cur/linux/tools/testing/selftests/net/lib/py/utils.py", line 77, in __init__ # Exception| raise Exception("Did not receive ready message") # Exception| Exception: Did not receive ready message Rename _get_unused_cpus() to _get_unused_rps_cpus() and cap the CPU search range to RPS_MAX_CPUS. Reviewed-by: Nimrod Oren Signed-off-by: Gal Pressman Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260210093110.1935149-1-gal@nvidia.com Signed-off-by: Jakub Kicinski

selftests: drivers: net: hw: Modify toeplitz.c to poll for packets

2026-02-11T10:10:20Z

Prior to this the receiver would sleep for the configured timeout, then attempt to receive as many packets as possible. This would result in a large burst of packets, and we don't necessarily need that many samples. The tests now run faster. Before ok 12 toeplitz.test.rps_udp_ipv6 # Totals: pass:12 fail:0 xfail:0 xpass:0 skip:0 error:0 real 0m54.792s user 0m12.486s sys 0m10.887s After ok 12 toeplitz.test.rps_udp_ipv6 # Totals: pass:12 fail:0 xfail:0 xpass:0 skip:0 error:0 real 0m36.892s user 0m4.203s sys 0m8.314s Signed-off-by: Jakub Kicinski Signed-off-by: Dimitri Daskalakis Link: https://patch.msgid.link/20260207013018.551347-1-dimitri.daskalakis1@gmail.com [pabeni@redhat.com: whitespaces fixes] Signed-off-by: Paolo Abeni

selftests: drv-net: rss: validate min RSS table size

2026-02-03T01:06:00Z

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 Link: https://patch.msgid.link/20260131225454.1225151-1-kuba@kernel.org Signed-off-by: Jakub Kicinski

selftests: drv-net: rss_flow_label: skip unsupported devices

2026-01-30T02:35:12Z

The test_rss_flow_label_6only test case fails on devices that do not support IPv6 flow label hashing. Make it skip neatly, consistent with the behavior of the test_rss_flow_label case. Reviewed-by: Gal Pressman Signed-off-by: Nimrod Oren Link: https://patch.msgid.link/20260128090217.663366-1-noren@nvidia.com Signed-off-by: Jakub Kicinski

selftests: drv-net: toeplitz: accept bigger rss keys

2026-01-29T03:37:14Z

/proc/sys/net/core/netdev_rss_key got bigger (256 bytes instead of 52) Fixes: 37b0ea8fef56 ("net: expand NETDEV_RSS_KEY_LEN to 256 bytes") Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20260127174806.886561-1-edumazet@google.com Signed-off-by: Jakub Kicinski

selftests: drv-net: fix missing include in ncdevmem

2026-01-21T02:26:26Z

Commit ca9d74eb5f6a ("uapi: add INT_MAX and INT_MIN constants") recently removed some includes of limits.h in uAPI headers. ncdevmem.c was depending on them: ncdevmem.c: In function ‘ethtool_add_flow’: ncdevmem.c:369:60: error: ‘INT_MAX’ undeclared (first use in this function) 369 | if (endptr == id_start || flow_id < 0 || flow_id > INT_MAX) | ^~~~~~~ ncdevmem.c:77:1: note: ‘INT_MAX’ is defined in header ‘’; did you forget to ‘#include ’? Reviewed-by: Mina Almasry Link: https://patch.msgid.link/20260120180319.1673271-1-kuba@kernel.org Signed-off-by: Jakub Kicinski

selftests: drv-net: extend HW timestamp test with ioctl

2026-01-21T02:21:27Z

Extend HW timestamp tests to check that ioctl interface is not broken and configuration setups and requests are equal to netlink interface. Some linter warnings are disabled because of ctypes classes. Reviewed-by: Kory Maincent Signed-off-by: Vadim Fedorenko Link: https://patch.msgid.link/20260116062121.1230184-2-vadim.fedorenko@linux.dev Signed-off-by: Jakub Kicinski

Merge tag 'net-queue-rx-buf-len-v9' of https://github.com/isilence/linux

2026-01-21T02:10:04Z

Pavel Begunkov says: ==================== Add support for providers with large rx buffer Many modern NICs support configurable receive buffer lengths, and zcrx and memory providers can use buffers larger than 4K to improve performance. When paired with hw-gro larger rx buffer sizes can drastically reduce the number of buffers traversing the stack and save a lot of processing time. It also allows to give to users larger contiguous chunks of data. Single stream benchmarks showed up to ~30% CPU util improvement. E.g. comparison for 4K vs 32K buffers using a 200Gbit NIC: packets=23987040 (MB=2745098), rps=199559 (MB/s=22837) CPU %usr %nice %sys %iowait %irq %soft %idle 0 1.53 0.00 27.78 2.72 1.31 66.45 0.22 packets=24078368 (MB=2755550), rps=200319 (MB/s=22924) CPU %usr %nice %sys %iowait %irq %soft %idle 0 0.69 0.00 8.26 31.65 1.83 57.00 0.57 This series adds net infrastructure for memory providers configuring the size and implements it for bnxt. It's an opt-in feature for drivers, they should advertise support for the parameter in the qops and must check if the hardware supports the given size. It's limited to memory providers as it drastically simplifies implementation. It doesn't affect the fast path zcrx uAPI, and the user exposed parameter is defined in zcrx terms, which allows it to be flexible and adjusted in the future. A liburing example can be found at [2] full branch: [1] https://github.com/isilence/linux.git zcrx/large-buffers-v8 Liburing example: [2] https://github.com/isilence/liburing.git zcrx/rx-buf-len * tag 'net-queue-rx-buf-len-v9' of https://github.com/isilence/linux: io_uring/zcrx: document area chunking parameter selftests: iou-zcrx: test large chunk sizes eth: bnxt: support qcfg provided rx page size eth: bnxt: adjust the fill level of agg queues with larger buffers eth: bnxt: store rx buffer size per queue net: pass queue rx page size from memory provider net: add bare bone queue configs net: reduce indent of struct netdev_queue_mgmt_ops members net: memzero mp params when closing a queue ==================== Link: https://patch.msgid.link/ Signed-off-by: Jakub Kicinski