Jakub Kicinski [off-list ref] writes:
On Wed, 11 Oct 2023 17:57:38 +0200 Arnd Bergmann wrote:
quoted
The .ndo_do_ioctl() netdev operation used to be how one communicates
with a network driver from userspace, but since my previous cleanup [1],
it is purely internal to the kernel.
Removing the cops appletalk/localtalk driver made me revisit the
missing pieces from that older series, removing all the unused
implementations in wireless drivers as well as the two kernel-internal
callers in the ieee802154 and appletalk stacks.
One ethernet driver was already merged in the meantime that should
have used .ndo_eth_ioctl instead of .ndo_do_ioctl, so fix that as well.
With the complete removal, any future drivers making this mistake
cause build failures that are easier to spot.
[1] https://lore.kernel.org/netdev/20201106221743.3271965-1-arnd@kernel.org/ (local)
Kalle, Johannes, do these apply for you?
I'm getting a small conflict on patch 8 and bigger one on patch 9.
Yes, 'git am -3' was able to solve the conflicts automatically and these
do apply to wireless-next.
If this applies for you maybe you can take it and flush out
wireless-next soon after?
Ok, we'll do that. Does next Wednesday sound soon enough? :)
--
https://patchwork.kernel.org/project/linux-wireless/list/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches