11:50 <NishanthMenon> just a headsup for folks https://lore.kernel.org/all/20240819204352.1423727-1-jm@ti.com/ -> we might have a push to go with upstream default as initramfs - I think most of the downstream distros already use initramfs to an extent.. but if folks have strong opinions, please chime in as necessary.

14:55 <javierm> NishanthMenon: it's too arbitrary IMO, I tried to enable a few options (as modules) needed for fedora and that got nacked by krzk

14:55 <javierm> https://lore.kernel.org/lkml/20240221141350.3740488-1-javierm@redhat.com/

14:56 <javierm> so it seems that are systems (e.g: android mentioned there) or filesystems (e.g: ext4) that have more weight than others

15:15 <NishanthMenon> yeah.. i was trying hard to tiptoe around biases ;) - thanks for chiming in the #armlinux channel

15:18 <javierm> NishanthMenon: I've been thinking about writing a tool (thought about naming it "kernelly" https://www.merriam-webster.com/dictionary/kernelly) to manage a set of defconfigs

15:19 <javierm> I've a local git repo that I use to manage my per board defconfig but is still a lot of manual work

15:20 <NishanthMenon> javierm: we had done something ;) https://git.ti.com/cgit/ti-linux-kernel/ti-upstream-tools/tree/config/defconfig_builder.sh https://git.ti.com/cgit/ti-linux-kernel/ti-upstream-tools/tree/config/defconfig_map.txt - we are lousy for nice naming though ;) -> but it kinda sets a reuse model with base config and a bunch of fragments.. there was some desire to discuss this in upstream, but we lost focus down the road

15:21 <javierm> NishanthMenon: interesting, thanks for sharing. I also though about posting a fedora.fragment but then lost interest due all the arguing

15:21 <javierm> https://lore.kernel.org/lkml/87v86h7a7l.fsf@minerva.mail-host-address-is-not-set/

15:22 <NishanthMenon> yeah - kinda sucks to see distros treated as second grade citizens in kernel :(

15:22 <javierm> yeah... and as mentioned the fedora defconfig for obvious reasons takes ages to build

15:22 <javierm> so every distro developer ends having cut down versions that they maintain, a lot of duplicated effort...

15:23 <javierm> NishanthMenon: anyways, thanks a lot for starting that discussion

15:23 <NishanthMenon> np. we had a similar "choice" challenge when we looked at https://git.ti.com/cgit/ti-linux-kernel/ti-upstream-tools/tree/config/containers.cfg -> do you enable docker? podman? lxc? what does containers configuration mean?

15:24 <NishanthMenon> i hope this translates to some interesting LPC side discussions.. (fingers crossed)

15:24 <javierm> NishanthMenon: hopefully, I'm not attending LPC this year but mripard from my team is

15:24 <NishanthMenon> aah ok.. for my sins, I guess I get to see Vienna next month :D

15:26 <javierm> :D