#archlinux-ports | Logs for 2025-11-26

Back
[00:02:21] <bschnei> FWIW, I have documented some of my own personal journey here: https://bens.haus a lot of things that I've learned I've written here and some of it should migrate to https://ports.archlinux.page so people can actually find it :)
[00:02:22] <phrik> Title: DiY Router | bens.haus (at bens.haus)
[00:47:09] -!- marmis has quit [Quit: Bye!]
[00:48:24] -!- marmis has joined #archlinux-ports
[04:35:03] -!- slainsavage has quit [Quit: WeeChat 4.7.1]
[04:41:13] -!- hcmb_ has joined #archlinux-ports
[04:41:13] hcmb is now known as Guest3112
[04:41:13] -!- Guest3112 has quit [Killed (tungsten.libera.chat (Nickname regained by services))]
[04:41:13] hcmb_ is now known as hcmb
[05:08:58] -!- ismailarilik has joined #archlinux-ports
[05:09:28] -!- ismailarilik has quit [Changing host]
[05:09:29] -!- ismailarilik has joined #archlinux-ports
[05:34:52] -!- slainsavage has joined #archlinux-ports
[05:41:25] -!- ismailarilik has quit [Ping timeout: 250 seconds]
[07:30:59] <slainsavage> solskogen|M: did you compile manage to finish :-)
[08:48:47] <mkurz|M> You might remember that I added the aarch64 sources to the obsidian package. Turns out we have a problem now: If we do not have aarch64 in the aarch array of the PKGBUILD, the checksums will not get updated, more here: https://gitlab.archlinux.org
[08:48:49] <phrik> Title: Add back aarch64 checksums (!6) · Merge requests · Arch Linux / Packaging / Packages / obsidian · GitLab (at gitlab.archlinux.org)
[08:49:39] <solskogen|M> mkurz|M: yeah, I know. That's why I've been advocating to allow that for certain packages.
[08:55:28] <mkurz|M> So should I add aarch64 to the array in my new merge request?
[08:56:01] <solskogen|M> That will probably be rejected, but you can try. You just have to argue well enough for it, I suppose.
[08:56:57] <solskogen|M> There's a alternative, and that is to have both files in the source array.
[08:59:21] <mkurz|M> hmm
[09:00:25] <slainsavage> solskogen|M: hi, did you manage to have time to check the dbeaver package?
[09:00:41] <solskogen|M> Oh, build is STILL going
[09:01:25] <solskogen|M> python-aotriton is at [135825/163981]
[09:01:50] * jelle shall ruin your build times next month
[09:01:55] <slainsavage> are there any other packages that you would like me to look at?
[09:02:03] <danct12|M> i'd like to help with ^ too
[09:02:20] <danct12|M> is there a list of packages that failed to build so anyone can look at?
[09:03:02] <solskogen|M> gstreamer and vlc
[09:03:37] <danct12|M> thats it!?
[09:03:58] <solskogen|M> I've HACKED it together ealier. but haven't made a patch for a MR
[09:04:14] <slainsavage> I will start taking a look at gstreamer shortly
[09:04:49] <slainsavage> solskogen|M: could you maybe pass the hack on?
[09:05:11] <solskogen|M> No. It was way to ugly.
[09:05:29] <slainsavage> all good :-)
[09:06:49] <slainsavage> just finishing up some NAXSI stuff for work, then will look at gstreamer
[09:08:07] <danct12|M> i probably should get my orange pi 5 back up again today
[09:08:24] <danct12|M> maybe i'll make a guide on how to set it up for new users too
[09:08:46] <danct12|M> kinda want to setup fex-emu so i can run steam and games too
[09:10:10] <slainsavage> danct12|M: if you don't mind me asking, what time zone are you in. I am in GMT+2
[09:10:57] <danct12|M> a few hours up ahead
[09:16:21] <solskogen|M> slainsavage: Oslo, Norway here!
[09:17:47] <slainsavage> solskogen|M: Cape Town, South Africa
[09:19:39] <solskogen|M> linux-tools is also something I need help with. It's a package that I don't fully understand and I'm not even sure how much of it we need for aarch64.
[09:20:16] <jelle> you want bpf
[09:21:34] <jelle> https://src.fedoraproject.org does some annoying things :D
[09:21:35] <phrik> Title: Making sure you're not a bot! (at src.fedoraproject.org)
[09:21:49] <solskogen|M> We've got that. But I see that ALARM has other tools enabled that aren't in x86_64. The thing is to figure out what is supported on aarch64 and what's not.
[09:22:03] <kxxt|M> solskogen|M: Some split-packages from it are specific to x86, we remove them in riscv64 port: https://github.com
[09:22:04] <phrik> Title: archriscv-packages/linux-tools/riscv64.patch at master · felixonmars/archriscv-packages · GitHub (at github.com)
[09:22:51] <solskogen|M> kxxt: yeah, I'd like one PKGBUILD that is for all archs, not only aarch64 :-)
[09:24:10] <jelle> I think that requires some pacman hacking
[09:24:24] <solskogen|M> okay.. why?
[09:24:40] <jelle> well how would you do split packages for specific arches?
[09:25:10] <Antiz> Yeah that's not supported as of now
[09:25:30] <kxxt|M> jelle: if [[ $CARCH == "x86_64 ]]; then pkgname+=(...); fi
[09:25:55] <solskogen|M> pkgname_$arch+=(...) should also work
[09:26:13] <jelle> solskogen|M: does it?
[09:27:46] <jelle> quick look in makepkg tells me it should not
[09:28:10] <Antiz> https://gitlab.archlinux.org
[09:28:12] <phrik> Title: Project space for the Aarch64 Linux project (#699) · Issues · Arch Linux / infrastructure · GitLab (at gitlab.archlinux.org)
[09:28:59] <Antiz> 👆 This indicates that makepkg may support multi-arch split packages, but our infra does not
[09:29:41] <Antiz> So, if not on the makepkg/pacman side, it at least requires some hacking on the infra side I guess :
[09:29:44] <Antiz> :P*
[09:30:12] <jelle> no, that is different
[09:30:19] <jelle> this isn't mixing x86_64 and any packages
[09:30:52] <jelle> this is having one pkgbase=foo which provides multiple packagenames but not all are applicable for aarch64 or riscv64
[09:31:09] <jelle> https://gitlab.alpinelinux.org
[09:31:15] <jelle> alpine seems to just provide cpupower on aarch64
[09:31:36] <Antiz> Ah right, my bad.
[09:38:00] <solskogen|M> My bad as well - I was so sure pkgname_$arch would work.
[09:40:02] <solskogen|M> but the workaround kxxt has are used here: https://gitlab.archlinux.org
[09:40:03] <phrik> Title: PKGBUILD · aarch64 · Benjamin Schneider / linux · GitLab (at gitlab.archlinux.org)
[12:37:20] -!- nl6720 has quit []
[12:39:11] -!- nl6720 has joined #archlinux-ports
[13:35:23] -!- slainsavage has quit [Quit: WeeChat 4.7.1]