Re: [PATCH net v5 4/4] selftests: net: add devmem RX and TX netdev unregister tests

[netdev-bot+sashiko@kernel.org]

Thank you for your contribution! Sashiko AI review found more severe
issue(s) in other patches of this series. This mail is only an FYI, in case
you want to address the nit picks below when you respin.

Low priority / nit picks for your consideration:

- [Low] The patch adds CONFIG_UDMABUF=y to
  tools/testing/selftests/net/config, but the fragment does not pull in…

--

selftests: net: add devmem RX and TX netdev unregister tests

This adds devmem_bind_rx_unregister_check and
devmem_bind_tx_unregister_check to nl_netdev.py. They check that removing a
netdevsim device while a netlink socket holds an RX or TX devmem binding
detaches the dma_buf right away. The commit also adds the dma-buf, udmabuf
and devmem options the tests need to the net selftest config fragment.

> diff --git a/tools/testing/selftests/net/config b/tools/testing/selftests/net/config
> index d4416f8bdcfd1..7a89e2618d4c4 100644
> --- a/tools/testing/selftests/net/config
> +++ b/tools/testing/selftests/net/config

[ ... ]

> @@ -133,6 +136,7 @@ CONFIG_TEST_BLACKHOLE_DEV=m
>  CONFIG_TLS=m
>  CONFIG_TRACEPOINTS=y
>  CONFIG_TUN=y
> +CONFIG_UDMABUF=y

[Severity: Low]
Will CONFIG_UDMABUF=y actually get enabled from this fragment on a minimal
base config? UDMABUF in drivers/dma-buf/Kconfig has these dependencies:

config UDMABUF
	bool "userspace dmabuf misc driver"
	default n
	depends on DMA_SHARED_BUFFER
	depends on MEMFD_CREATE || COMPILE_TEST
	depends on MMU

CONFIG_SYNC_FILE=y takes care of DMA_SHARED_BUFFER. MEMFD_CREATE has no
default, though, and can only be set directly under EXPERT. Normally it
is turned on because TMPFS or HUGETLBFS selects it, and this fragment
asks for neither.

If the base config has MEMFD_CREATE=n and COMPILE_TEST=n, Kconfig will
quietly drop UDMABUF. Both new tests would then skip in _create_udmabuf()
and never reach the unregister path:

    if not os.path.exists("/dev/udmabuf"):
        raise KsftSkipEx("/dev/udmabuf is not available")

tools/testing/selftests/bpf/config lists CONFIG_TMPFS=y next to
CONFIG_UDMABUF=y, and tools/testing/selftests/drivers/net/hw/config has
CONFIG_HUGETLBFS=y. Should CONFIG_TMPFS=y be added here as well?

>  CONFIG_USER_NS=y
>  CONFIG_VETH=y
>  CONFIG_VLAN_8021Q=y

[ ... ]

-- 
Sashiko AI review · https://netdev-ai.bots.linux.dev/sashiko/#/patchset/20261030083539.777877-1-almasrymina%40google.com