Re: [PATCH 01/10 net-next v2] ipv6: convert CONFIG_IPV6 to built-in only and clean up Kconfigs

Hi Fernando,

On Thu, 12 Mar 2026 at 16:12, Fernando Fernandez Mancera
<[email protected]> wrote:
> On 3/12/26 4:02 AM, Jakub Kicinski wrote:
> > On Tue, 10 Mar 2026 16:34:24 +0100 Fernando Fernandez Mancera wrote:
> >> Maintaining a modular IPv6 stack offers image size and memory savings
> >> for specific setups, this benefit is outweighed by the architectural
> >> burden it imposes on the subsystems on implementation and maintenance.
> >> Therefore, drop it.
> >>
> >> Change CONFIG_IPV6 from tristate to bool. Remove all Kconfig
> >> dependencies across the tree that explicitly checked for IPV6=m. In
> >> addition, remove MODULE_DESCRIPTION(), MODULE_ALIAS(), MODULE_AUTHOR()
> >> and MODULE_LICENSE().
> >>
> >> This is also replacing module_init() by device_initcall(). It is not
> >> possible to use fs_initcall() as IPv4 does because that creates a race
> >> condition on IPv6 addrconf.
> >>
> >> Finally, modify the default configs from CONFIG_IPV6=m to CONFIG_IPV6=y
> >> except for m68k as according to the bloat-o-meter the image is
> >> increasing by 330KB~ and that isn't acceptable. Instead, disable IPv6 on
> >> this architecture by default. This is aligned with m68k RAM requirements
> >> and recommendations [1].
> >
> > AI has spotted:
> >
> >> diff --git a/arch/m68k/configs/amiga_defconfig b/arch/m68k/configs/amiga_defconfig
> >> index 31d16cba9879..de088071dde4 100644
> >> --- a/arch/m68k/configs/amiga_defconfig
> >> +++ b/arch/m68k/configs/amiga_defconfig
> >> @@ -64,7 +64,6 @@ CONFIG_NET_IPIP=m
> >>   CONFIG_NET_IPGRE_DEMUX=m
> >>   CONFIG_NET_IPGRE=m
> >>   CONFIG_NET_IPVTI=m
> >> -CONFIG_NET_FOU_IP_TUNNELS=y
> >>   CONFIG_INET_AH=m
> >
> > Is CONFIG_NET_FOU_IP_TUNNELS=y removed intentionally? This option
> > provides FOU/GUE encapsulation for IP tunnels and has 'depends on
> > NET_IPIP || NET_IPGRE || IPV6_SIT' as its Kconfig dependency. With IPv6
> > disabled, IPV6_SIT becomes unavailable, but CONFIG_NET_IPIP=m and
> > CONFIG_NET_IPGRE=m are both still present in the defconfig, so the
> > dependency remains satisfiable.
> >
> > Since CONFIG_NET_FOU_IP_TUNNELS has no 'default y', removing it from the
> > defconfig means FOU/GUE encapsulation for IP tunnels will be silently
> > disabled by default on m68k. The commit message describes only disabling
> > IPv6 on m68k, not removing IPv4 FOU tunnel support.
> >
>
> I noticed that when running
>
> ./scripts/config --disable CONFIG_IPV6
>
> for the m68k, the script was adding CONFIG_LWTUNNEL=y and CONFIG_NET_FOU=y.
>
> CONFIG_LWTUNNEL was selected by multiple IPV6 features. I do not think
> it makes sense to keep it for m68k given the information there is on
> http://www.linux-m68k.org/faq/platinfo.html.

Dunno about lwtunnel...

> CONFIG_NET_FOU was something IPV6_FOU required, probably it should be
> just dropped from the config instead of explicitly turn it off as it
> turns off FOU_IP_TUNNELS too. It will be selected by FOU_IP_TUNNELS too
> anyway.

... but CONFIG_NET_FOU seems to be just a gatekeeping symbol
for the tristate symbol IPV6_FOU, so FOU is still a module., which is good.



Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds