kernel/Documentation/networking/napi.rst, branch linux-rolling-stable

net: Extend NAPI threaded polling to allow kthread based busy polling

2025-11-04T02:11:40Z

Add a new state NAPI_STATE_THREADED_BUSY_POLL to the NAPI state enum to enable and disable threaded busy polling. When threaded busy polling is enabled for a NAPI, enable NAPI_STATE_THREADED also. When the threaded NAPI is scheduled, set NAPI_STATE_IN_BUSY_POLL to signal napi_complete_done not to rearm interrupts. Whenever NAPI_STATE_THREADED_BUSY_POLL is unset, the NAPI_STATE_IN_BUSY_POLL will be unset, napi_complete_done unsets the NAPI_STATE_SCHED_THREADED bit also, which in turn will make the kthread go to sleep. Signed-off-by: Samiullah Khawaja Reviewed-by: Willem de Bruijn Acked-by: Martin Karsten Tested-by: Martin Karsten Link: https://patch.msgid.link/20251028203007.575686-2-skhawaja@google.com Signed-off-by: Jakub Kicinski

docs: remove obsolete description about threaded NAPI

2025-09-01T20:09:08Z

Commit 2677010e7793 ("Add support to set NAPI threaded for individual NAPI") introduced threaded NAPI configuration per individual NAPI instance, however obsolete description that threaded NAPI is per device has remained. Remove the old description and clarify that only NAPI instances running in threaded mode spawn kernel threads by changing "Each NAPI instance" to "Each threaded NAPI instance". Signed-off-by: Kohei Enju Reviewed-by: Samiullah Khawaja Link: https://patch.msgid.link/20250829064857.51503-1-enjuk@amazon.com Signed-off-by: Jakub Kicinski

Add support to set NAPI threaded for individual NAPI

2025-07-15T01:02:37Z

A net device has a threaded sysctl that can be used to enable threaded NAPI polling on all of the NAPI contexts under that device. Allow enabling threaded NAPI polling at individual NAPI level using netlink. Extend the netlink operation `napi-set` and allow setting the threaded attribute of a NAPI. This will enable the threaded polling on a NAPI context. Add a test in `nl_netdev.py` that verifies various cases of threaded NAPI being set at NAPI and at device level. Tested ./tools/testing/selftests/net/nl_netdev.py TAP version 13 1..7 ok 1 nl_netdev.empty_check ok 2 nl_netdev.lo_check ok 3 nl_netdev.page_pool_check ok 4 nl_netdev.napi_list_check ok 5 nl_netdev.dev_set_threaded ok 6 nl_netdev.napi_set_threaded ok 7 nl_netdev.nsim_rxq_reset_down # Totals: pass:7 fail:0 xfail:0 xpass:0 skip:0 error:0 Signed-off-by: Samiullah Khawaja Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20250710211203.3979655-1-skhawaja@google.com Signed-off-by: Jakub Kicinski

documentation: networking: Add NAPI config

2025-02-15T01:09:19Z

Document the existence of persistent per-NAPI configuration space and the API that drivers can opt into. Update stale documentation which suggested that NAPI IDs cannot be queried from userspace. Signed-off-by: Joe Damato Acked-by: Jakub Kicinski Reviewed-by: Bagas Sanjaya Link: https://patch.msgid.link/20250213191535.38792-1-jdamato@fastly.com Signed-off-by: Jakub Kicinski

documentation: networking: fix spelling mistakes

2025-01-27T22:26:06Z

Fix a couple of typos/spelling mistakes in the documentation. Signed-off-by: Khaled Elnaggar Acked-by: Marc Kleine-Budde Reviewed-by: Przemek Kitszel Acked-by: Randy Dunlap Reviewed-by: Bagas Sanjaya Link: https://patch.msgid.link/20250123082521.59997-1-khaledelnaggarlinux@gmail.com Signed-off-by: Jakub Kicinski

tools: ynl: move python code to separate sub-directory

2025-01-09T20:53:27Z

Move python code to a separate directory so it can be packaged as a python module. Updates existing references in selftests and docs. Also rename ynl-gen-[c|rst] to ynl_gen_[c|rst], avoid dashes as these prevent easy imports for entrypoints. Signed-off-by: Jan Stancek Reviewed-by: Donald Hunter Link: https://patch.msgid.link/a4151bad0e6984e7164d395125ce87fd2e048bf1.1736343575.git.jstancek@redhat.com Signed-off-by: Jakub Kicinski

docs: networking: Describe irq suspension

2024-11-12T02:45:06Z

Describe irq suspension, the epoll ioctls, and the tradeoffs of using different gro_flush_timeout values. Signed-off-by: Joe Damato Co-developed-by: Martin Karsten Signed-off-by: Martin Karsten Reviewed-by: Sridhar Samudrala Reviewed-by: Bagas Sanjaya Link: https://patch.msgid.link/20241109050245.191288-7-jdamato@fastly.com Signed-off-by: Jakub Kicinski

doc: net: napi: Update documentation for napi_schedule_irqoff

2024-10-03T10:07:29Z

Since commit 8380c81d5c4f ("net: Treat __napi_schedule_irqoff() as __napi_schedule() on PREEMPT_RT"), napi_schedule_irqoff will do the right thing if IRQs are threaded. Therefore, there is no need to use IRQF_NO_THREAD. Signed-off-by: Sean Anderson Reviewed-by: Bagas Sanjaya Reviewed-by: Sebastian Andrzej Siewior Link: https://patch.msgid.link/20240930153955.971657-1-sean.anderson@linux.dev Signed-off-by: Paolo Abeni

docs: net: clarify the NAPI rules around XDP Tx

2023-07-22T01:51:37Z

page pool and XDP should not be accessed from IRQ context which may happen if drivers try to clean up XDP TX with NAPI budget of 0. Link: https://lore.kernel.org/r/20230720161323.2025379-1-kuba@kernel.org Signed-off-by: Jakub Kicinski

docs: networking: document NAPI

2023-03-24T02:47:40Z

Add basic documentation about NAPI. We can stop linking to the ancient doc on the LF wiki. Link: https://lore.kernel.org/all/20230315223044.471002-1-kuba@kernel.org/ Reviewed-by: Bagas Sanjaya Reviewed-by: Toke Høiland-Jørgensen Acked-by: Pavel Pisa # for ctucanfd-driver.rst Reviewed-by: Tony Nguyen Reviewed-by: Florian Fainelli Reviewed-by: Stephen Hemminger Reviewed-by: Randy Dunlap Link: https://lore.kernel.org/r/20230322053848.198452-1-kuba@kernel.org Signed-off-by: Jakub Kicinski