09:03 <chadmed_> jannau: where would you put a hypothetical ebuild for asahi-scripts? im thinking in sys-apps since theres no sys-util

09:03 <chadmed_> (keeping in mind we cant make up categories because we eventually want this stuff upstreamed)

09:13 <jannau> chadmed_: maybe sys-boot or imo less fitting app-admin. one of those 3

09:15 <chadmed_> app-admin sounds good actually since im going to bring the fw update script with it

09:15 <chadmed_> that way we dont have to host files in a bunch of random directories for the ebuilds, we just depend on asahi-scripts

09:16 <Sobek[m]> Oh, so this is going to be an asahi overlay, which should make gentoo asahi easy?

09:18 <chadmed_> https://github.com/chadmed/asahi-gentoosupport https://github.com/chadmed/asahi-overlay <-- way ahead of you ;)

09:18 <jannau> Sobek[m]: it's already there: https://github.com/chadmed/asahi-gentoosupport

09:18 <chadmed_> the asahi wiki has a page on installing it all, im just bringing us up to speed with upstream changes that ive neglected to track for the past week

09:19 <Sobek[m]> And you are just improving the way it is packaged :)

09:20 <jannau> chadmed_: I think it makes sense to ship the scripts with their packages but I see that it makes updating them very inconvenient. maybe we should fetch them with the m1n1/asahi-firmware ebuild instead of shipping it a file

09:21 <Sobek[m]> What's kmod, and why does it need to be built using gcc (and not clang) ?

09:22 <jannau> package handling kernel modules loading

09:23 <jannau> chadmed_: app-admin sounds fitting but the other packages in app-admin are not good fit

09:27 <Sobek[m]> Other question on that overlay, why is kile in it (does it require patches on arm64 ?)

09:30 <jannau> btw I wouldn't invest too much time in initrd firmware handling. It will change soon. initrd will get the firmware via a second cpio archive and then has to provide it to the rootfs before switching root

09:37 <chadmed_> Sobek[m]: upstream have it masked for arm64 but i need it for uni. the philosophy behind the overlay at the moment is "stuff that makes these machines work nicely"

09:37 <chadmed_> kile makes these machines nicer to use and therefore i will have it there until i can be bothered pushing the keyword upstream

09:38 <chadmed_> i dont have time for that particular rigmarole at the moment

09:39 <chadmed_> jannau: sys-apps does seem like the more logical home for it

09:40 <jannau> chadmed_: not a fan of "app-editor/kile **" in package.accept_keywords?

09:40 <jannau> chadmed_: sys-apps is fine for me. if upstream disagrees moving it to another category isn't really an issue

09:41 <Sobek[m]> jannau: Yeah, I wonder why you need an overlay package if it is just masked ? Specific patches needed ?

09:43 <jannau> I have quite a few packages with ** I haven't yet bothered to report as working on arm64

09:45 <chadmed_> jannau: now i could pretend i forgot about accept_keywords and say that i added it to the overlay so that it could benefit everyone, but that would be a lie

09:45 <chadmed_> its been a very annoying and rough couple of weeks youll have to forgive me

09:50 <jannau> np, no harm done

09:53 <jannau> chadmed_: have you seen the pull request from fedora in the asahi-scripts repo. we should see if we can use the dracut module

09:57 <chadmed_> yeah i like that idea, once its all sorted out we can have a play around with it

09:57 <chadmed_> im just getting a patch ready for update-m1n1 since it looks for -ARCH dtbs, we want it to install already looking in /usr/src/linux/arch/arm64/boot/dts/apple

19:35 <espo> @chadmed[m] i created a pull request for the m1n1-1.1.5-r2 ebuild :) you moved the asahi scripts but the ebuild still uses app-admin :)