#archlinux-ports | Logs for 2026-04-22

Back
[00:07:04] -!- filmroellchen has quit [Quit: filmroellchen]
[00:38:12] -!- Charon77 has joined #archlinux-ports
[00:52:52] <Charon77> I'm playing with packaging, just stumbled upon `repo-elephant`. Made my day :)
[01:16:24] -!- marmis has joined #archlinux-ports
[01:18:24] -!- ubitux has quit [Ping timeout: 246 seconds]
[01:20:53] -!- ubitux has joined #archlinux-ports
[01:24:54] <fermino> NICE xD
[01:28:17] <fermino> Solskogen, any chance `latest` tar+sig symlinks could be provided for the bootstrap tarballs at https://arch-linux-repo.drzee.net ? Just so that I can automate the image creation without having to parse and sort the index :)
[01:28:18] <phrik> Title: Index of arch/tarballs/os/aarch64/ (at arch-linux-repo.drzee.net)
[01:36:25] <fermino> (Not that I think it's a good idea to parse the index xD)
[01:37:24] <Charon77> fermino: what are you trying to do?
[01:38:00] <fermino> I'm making something similar to arch-boxes but hopefully a little bit prettier
[01:38:30] <fermino> The idea is to have different system images defined in a declarative style (currently a subset of HCL)
[01:39:38] <fermino> So rn I'm making a "bootstrap" module that takes care of unpacking the tarball from the repo in a partition
[01:42:13] <Charon77> I see.
[01:42:14] <Charon77> I still don't get what you mean by tar+sig symlinks. Are you trying to get the list of packages included in the tarball?
[01:44:03] <fermino> Oh no, I mean a latest symlink for the .tar.zst and one for the signature; I would like to avoid having to update the tarball url every time there's a tarball update
[01:46:22] <Charon77> I see! That would be useful!
[02:20:43] -!- TheDcoder has quit [Remote host closed the connection]
[02:20:59] -!- TheDcoder has joined #archlinux-ports
[03:19:57] -!- hcmb_ has joined #archlinux-ports
[03:19:57] hcmb is now known as Guest7305
[03:19:58] hcmb_ is now known as hcmb
[03:21:05] -!- Guest7305 has quit [Ping timeout: 252 seconds]
[04:00:37] -!- TheDcoder has quit [Read error: Connection reset by peer]
[04:01:16] -!- TheDcoder has joined #archlinux-ports
[04:20:42] -!- TheDcoder has quit [Read error: Connection reset by peer]
[04:21:16] -!- TheDcoder has joined #archlinux-ports
[06:07:59] -!- solsTiCe4 has joined #archlinux-ports
[06:10:19] -!- solsTiCe has quit [Ping timeout: 276 seconds]
[06:10:19] solsTiCe4 is now known as solsTiCe
[06:18:23] -!- drathir_tor has quit [Remote host closed the connection]
[06:18:47] -!- drathir_tor has joined #archlinux-ports
[06:25:43] -!- TheDcoder has quit [Read error: Connection reset by peer]
[06:26:19] -!- TheDcoder has joined #archlinux-ports
[06:30:43] -!- solsTiCe has quit [Ping timeout: 265 seconds]
[06:58:30] -!- drathir_tor has quit [Remote host closed the connection]
[07:03:49] -!- drathir_tor has joined #archlinux-ports
[07:12:40] -!- marmis7 has joined #archlinux-ports
[07:14:13] -!- marmis has quit [Ping timeout: 265 seconds]
[07:14:14] marmis7 is now known as marmis
[07:55:55] -!- TheDcoder has quit [Read error: Connection reset by peer]
[07:56:21] -!- TheDcoder has joined #archlinux-ports
[08:16:00] -!- TheDcoder has quit [Read error: Connection reset by peer]
[08:16:29] -!- TheDcoder has joined #archlinux-ports
[08:23:48] <solskogen|M> fermino: I'll ask @drzee
[08:32:34] <solskogen|M> AGX?
[08:34:17] <solskogen|M> oh, I see. Is it reported upstream?
[08:38:14] <solskogen|M> the one I built was a unmodified PKGBUILD from x86_64, you might want to take a look
[08:44:30] -!- Antiz has quit [Quit: The Lounge - https://thelounge.chat]
[08:49:48] -!- filmroellchen has joined #archlinux-ports
[08:58:31] -!- meijin007 has joined #archlinux-ports
[08:59:48] <meijin007> how "hard" (or more accurately, what it will take) to make the arm64 port work on RPI4?
[09:01:04] <solskogen|M> It requires a big amount on work on my part, but bschnei has a repo which should work. His repo isn't as big as the one on @drzee
[09:01:49] <meijin007> can I ask why it such a difference?
[09:02:02] <solskogen|M> Historic reasons.
[09:03:03] <meijin007> :( ok
[09:03:24] <meijin007> it's just the prices for sbc are creazy now
[09:03:48] <solskogen|M> That said, it is possible to have both armv8.2-a packages and armv8-a packages in a repo. That's not an issue. Heck, even @drzee can host them.
[09:04:43] <solskogen|M> But I'm not gonna spend my time compiling and fixing any problems that might gonna show up
[09:05:07] <meijin007> Is it a matter of specifying a more specific arch in pacman.conf?
[09:05:27] <solskogen|M> and makepkg.conf - and in how you compile gcc.
[09:05:33] <meijin007> yes, that's makes sense :\
[09:06:02] <solskogen|M> and you need a kernel (that's the easy part)
[09:06:19] <solskogen|M> and you need to recompile everything
[09:08:14] <solskogen|M> if anyone is up for the task, you can just use the armv8.2-a packages for bootstrapping, as long as the build machine can use them. you don't have to bootstrap from scratch.
[09:11:51] <Charon77> I'm doing this right now. I have a server running on armv8.2-a but compiling for armv8 (possibly +crc)
[09:11:52] <Charon77> I populated the packages from bschnei. I have firefox and it runs on my armv8 chromebook :D
[09:13:14] <Charon77> I'm currently putting it on oracle bucket, but would love if we can have a shared location to put the packages to
[09:15:41] -!- cjc7373 has quit [Remote host closed the connection]
[09:16:05] -!- cjc7373 has joined #archlinux-ports
[09:18:01] -!- Solskogen has quit [Read error: Connection reset by peer]
[09:25:44] -!- Guest2054 has joined #archlinux-ports
[09:31:51] <Charon77> meijin007: bschnei's repo may not be as big, but it's still good enough! (I use it right now)
[09:32:35] <Charon77> I've got window manager (niri-wm), wayland stuffs working out of the box
[10:39:08] -!- Antiz has joined #archlinux-ports
[11:06:20] <linkmauve> > (possibly +crc)
[11:06:21] <linkmauve> You understand that this means it won’t run on a Raspberry Pi?
[11:06:58] <linkmauve> And all libraries for which this is a significant increase in performances will pick it at runtime anyway.
[11:08:07] <Charon77> There's a package (dpdk) that bschnei mentioned won't compile without +crc. When I tried building it, I couldn't. Other packages doesn't seem to need it
[11:08:21] <Charon77> I didn't know Raspberry Pi don't have crc
[11:08:43] <solskogen|M> I think they do.
[11:44:39] <solskogen|M> with regards to dpdk, the problem here is LTO. If you disable LTO it builds fine.
[12:11:27] -!- marmis has quit [Quit: Bye!]
[12:12:54] -!- marmis has joined #archlinux-ports
[12:21:54] -!- marmis7 has joined #archlinux-ports
[12:23:19] -!- marmis has quit [Read error: Connection reset by peer]
[12:23:19] marmis7 is now known as marmis
[12:53:39] <cjc7373> I successfully built ghc using official prebuilt binary as the bootstrapper. I suppose I can write a documentation and then solskogen can build it and all haskell packages and push them to the repo?
[12:58:18] <solskogen|M> yes!
[12:58:40] <anthraxx|M> cjc7373: if you can document that, that would be amazing. you can open an MR against https://gitlab.archlinux.org
[12:58:41] <phrik> Title: Arch Linux / Ports / docs · GitLab (at gitlab.archlinux.org)
[12:59:25] <solskogen|M> last time I tried, I wasn't even able to compile ghc with the current gcc.
[13:15:19] <Charon77> solskogen|M:
[13:15:20] <Charon77> > That said, it is possible to have both armv8.2-a packages and armv8-a packages in a repo. That's not an issue. Heck, even @drzee can host them.
[13:15:20] <Charon77> Any chance I could help with this? I have a cloud VM that can compile packages and I'm willing to debug armv8 issues because that's what I use. Having a proper place to host would be nice.
[13:16:26] <solskogen|M> I'll ask @drzee to join IRC, Charon77 :-)
[13:27:01] <Charon77> nice. In the meantime I'm having a bucket with (mostly) bschnei and my packages
[13:32:29] <cjc7373> solskogen|M: check https://gitlab.archlinux.org :)
[13:32:30] <phrik> Title: bootstrap haskell compiler (ghc) (#11) · Issues · Arch Linux / Ports / AArch64 / project-management · GitLab (at gitlab.archlinux.org)
[13:34:43] <solskogen|M> cjc7373: okay, so it's not finished? IIRC hadrian is a dependency of ghc
[13:35:14] <Charon77> solskogen|M: should I do a writeup about compiling firefox with the llvm issue?
[13:35:52] <cjc7373> well the build order is for ghc, not only for hadrian
[13:36:19] <solskogen|M> Charon77: if you wish, but I found another workaround that seems to do the trick.
[13:37:15] <solskogen|M> https://gitlab.archlinux.org wont build without hadrian, I mean.
[13:37:16] <phrik> Title: Arch Linux / Packaging / Packages / ghc · GitLab (at gitlab.archlinux.org)
[13:38:13] <cjc7373> hadrian is also in the build list
[13:38:27] <Charon77> the headless workaround?
[13:38:37] <solskogen|M> Charon77: yes.
[13:39:00] <solskogen|M> cjc7373: Oh, I see it now. Let me try.
[13:42:12] <cjc7373> I forget to mention that you may also need --nocheck, checkdepends are not considered
[13:42:51] <solskogen|M> That's fine, for now :)
[14:01:45] <Charon77> https://gitlab.archlinux.org
[14:01:47] <phrik> Title: Firefox build note (#12) · Issues · Arch Linux / Ports / AArch64 / project-management · GitLab (at gitlab.archlinux.org)
[14:04:02] <solskogen|M> haskell-build in progress: https://arch-linux-repo.drzee.net
[14:04:03] <phrik> Title: Arch Linux AArch64 Build Report (at arch-linux-repo.drzee.net)
[14:06:43] Guest2054 is now known as Solskogen
[14:12:59] <Charon77> that is a lot of packages
[14:13:38] <Charon77> wait, why does it say aborted
[14:14:16] <solskogen|M> Because I stopped it
[14:14:52] <solskogen|M> I forgot to not install checkdepends
[14:15:11] <Charon77> :D
[14:16:23] <solskogen|M> due to the fact that a lot of packages have dependencies in checkdepends and not makedepends I need to install them seperately
[14:24:35] -!- meijin007 has quit [Quit: Client closed]
[14:35:21] -!- Antiz has quit [Ping timeout: 248 seconds]
[14:44:32] -!- Antiz has joined #archlinux-ports
[15:06:45] <bschnei> maybe what makes sense to do (have to check with drzee) is take the packages I already have in my personal repo and move that all to drzee.net. there's really no need to recompile everything for v8 if i'm only missing 1000 packages and we're getting reports that things actually work :)
[15:08:03] <bschnei> the real question remains how to facilitate collaboration for those in the community who can/do want to work on v8. does hosting the packages on drzee.net actually solve that problem? i'm not sure....
[15:16:22] <Charon77> Let's say I see a package is missing. I build it, then I send to drzee.net, which will add it to index
[15:16:23] <Charon77> Wouldn't that be sufficient?
[15:20:47] <bschnei> I wish :) you know that whole "don't do partial upgrades!" mantra. Ya, you wouldn't want to publish a single package that may be ahead/behind all of the other packages that might depend on it. That's how you break everyone's devices :)
[15:21:30] <bschnei> That's where the staging repos generally come into play. You'll see _sets_ of packages getting rebuilt before they actually get published
[15:23:20] <bschnei> Single package work is still highly valuable, but releasing a built package is not always as straightforward
[15:24:54] <Charon77> For missing package, I think it should be fine, right?
[15:24:55] <Charon77> The issue is when upgrading a package, the rest of the dependency chain needs to be rebuild
[15:25:31] -!- TheDcoder has quit [Remote host closed the connection]
[15:25:39] <bschnei> Ya but what packages are you using to build that missing package in the first place?
[15:25:53] <bschnei> Ones from my personal repo that have fallen behind?
[15:26:12] <bschnei> Ones you've built yourself that are ahead?
[15:26:28] -!- TheDcoder has joined #archlinux-ports
[15:26:29] <Charon77> I'm using solskogen's armv8.2-a to build armv8 packages
[15:27:19] <bschnei> Right which are very up to date. So if you then publish a package built against very up to date packages into a repo with a bunch of old dependencies: breakage.
[15:28:31] <bschnei> Here's the good news. You don't really have to worry about that at the moment. There are so many packages that just don't build and if you can resolve the issues, I can add them to my personal repo which...isn't anywhere near as up to date as v8.2 but it's also not terrible
[15:30:00] <bschnei> Then you can sync the v8 repos from my personal site to continue testing, finding what's missing, etc.
[15:30:43] <bschnei> In other words, if you are willing to work on v8 packaging, I can help with publishing the work/managing the repos since I already have them
[15:31:24] <bschnei> like solskogne, I don't have much capacity/motivation for actually working on v8 package issues generally. the packages I use on my router all work and that's good enough for me
[15:32:02] <bschnei> But I'm happy to contribute hosting/repo mgmt to the effort
[15:33:33] <Charon77> I think relying on just you would be bottlenecking on the building device. I can contribute computing power too, and so does others. I feel like what's needed is the managing people's compilation effort
[15:34:27] <bschnei> We're saying the same thing. I don't care if you send me a .pkg. or you ask me to build it. I've already got the repo up (though it's not as robust as drzee.net) and can publish the work
[15:34:55] <bschnei> 99% of packages do not take that long to build. that's not the bottleneck
[15:36:41] <bschnei> In practice a simple DM to me that says "hey, X is building for me on v8 but is missing, pls add" is all you really gotta do
[15:37:50] <Charon77> Alright, let's try it now (but here)
[15:38:30] <bschnei> If you give me a min, I can get you an up to date list of packages that i don't have in my repos. it is only partially filtered for packages that don't make sense and haskell related stuff, but if it's helpful I can post
[15:38:32] <Charon77> firefox: It builds if we use headless (edit PKGBUILD, MOZ_HEADLESS=1), or if we patch llvm https://gitlab.archlinux.org
[15:38:33] <phrik> Title: [Ports:AArch64] fix AArch64AsmPrinter (!8) · Merge requests · Arch Linux / Packaging / Packages / llvm · GitLab (at gitlab.archlinux.org)
[15:40:06] <Charon77> KDE stuffs: try installing dolphin, for me it's missing these dependencies: kdsoap kdsoap-ws-discovery-client kio-extras smb4k plasma-workspace
[15:40:06] <Charon77>
[15:40:06] <Charon77> I was able to build all of them
[15:40:13] <Charon77> (pretty quickly too)
[15:40:24] <bschnei> re: firefox let's wait until that's merged/released. Otherwise we have a versioning/reproducibility problem where the version/tag on gitlab doesn't match what we actually have in our repos
[15:42:18] <Charon77> So we don't use forked PKGBUILD
[15:42:19] <Charon77>
[15:42:20] <Charon77> > let's wait until that's merged
[15:42:20] <Charon77> can I ask for review :D
[15:43:29] <bschnei> I don't unless it is something that can't/won't be merged upstream yet. I'm not interested in maintaining hundreds of forks...I've got quite a few already lol
[15:45:29] <bschnei> I'd give Peter some time. It's only been two days. I have an MR open for 7 months...sometimes things need to bake for a bit and/or if upstream already has a fix and they will release anyway in a week, then it doesn't make sense to muck up PKGBUILD with patches
[15:46:24] <Charon77> btw do you know about this: https://ports.archlinux.page
[15:46:25] <Charon77> > All secondary architecture related merge-requests must be flagged with the new pkg::ports GitLab label.
[15:46:25] <Charon77>
[15:46:25] <Charon77> I don't think I can put label on gitlab. permission?
[15:46:25] <phrik> Title: Arch Linux Ports (at ports.archlinux.page)
[15:47:10] <bschnei> I think that was something they were going to try to figure out how to automate to make it easier on everyone, but it never came to pass. I think it's safe to disregard for now
[15:47:24] <bschnei> kdsoap added
[15:51:05] -!- alkersan has quit [Ping timeout: 244 seconds]
[15:54:16] -!- alkersan has joined #archlinux-ports
[16:01:22] <Charon77> should I really tell you every single missing packages? :)
[16:01:23] <Charon77>
[16:01:23] <Charon77> I mean, some packages are just not built yet, while other packages need more investigation
[16:02:54] -!- Antiz has quit [Quit: The Lounge - https://thelounge.chat]
[16:05:01] <Charon77> What I'm coming from is: I have time and patience, I have a VM, how can I best contribute to the project.
[16:05:14] -!- Antiz has joined #archlinux-ports
[16:09:02] <bschnei> Investigate the ones that aren't building
[16:10:04] <solskogen|M> keep in mind that building isn't everything. A package might build, but that doesn't mean that it works. So testing stuff is quite important :)
[16:10:21] <Charon77> ah true
[16:10:28] <Charon77> sure, just give me the list (or have it as work item somewhere)
[16:10:30] <bschnei> https://paste.rs
[16:10:40] <bschnei> ^ these are missing from my (v8) repos right now
[16:10:53] <bschnei> missing could mean i just haven't even tried yet to build it (like the KDE packages you mentioned)
[16:11:05] <bschnei> but a lot of them just don't build period. not on aarch64, not on x64.
[16:11:31] <bschnei> same may be irrelevant
[16:11:33] <bschnei> some*
[16:11:51] <solskogen|M> bschnei: like aarch64-linux-gnu-gcc :-)
[16:11:55] <bschnei> exactly
[16:12:26] <bschnei> so ignore things that are obviously n/a, but you can see there is a still quite a bit of investigating to do
[16:14:00] <Charon77> if I can build it on armv8-2.a VM with armv8-2.a packages (but targetting armv8-a) is that enough?
[16:14:10] <bschnei> yup
[16:14:17] <Charon77> also what's your target? armv8-a? +crc?
[16:14:28] <bschnei> and with things like dolphin above, just lmk dolphin and i can look through the deps myself
[16:14:44] <solskogen|M> I hope you're not using my gcc. because that will default on armv8.2-a.
[16:14:46] <bschnei> just armv8-a, no +crc at the moment...also not super well understood (by me)
[16:15:16] <bschnei> ^ah yes be mindful of what solskogen said
[16:15:43] <bschnei> but again, it's better to just let me know something builds for you and i can rebuild and release
[16:16:07] <bschnei> so i've got one for ya
[16:16:22] <Charon77> I override CLFAGS on makepkg.conf
[16:16:34] <solskogen|M> Charon77: Not enough
[16:16:42] <Charon77> :<
[16:17:01] <bschnei> dolphin needs qt6-base...qt6-base needs libb2. official libb2 doesn't build without modifications
[16:17:04] <solskogen|M> some packages clear our CFLAGS
[16:17:48] <bschnei> go to libb2 for gitlab and see there are no forks or MRs
[16:18:08] <bschnei> error it throws is configure: error: Compiler does not know -msse2.
[16:18:55] <bschnei> which obviously is not an ARM instruction set
[16:20:01] <bschnei> ideally upstream's configure script is smarter. not sure why it isn't in this case. but the point is...this has absolutely nothing at all to do with v8 vs v8.2
[16:20:21] <bschnei> and that is probably 90% of what you will find
[16:20:45] <bschnei> which is why the discussion about whether we support v8 vs v8.2 is very premature. there's so much we don't understand
[16:20:49] <linkmauve> How did that build on 8.2?
[16:20:52] <bschnei> (or at least I don't)
[16:24:55] <cjc7373> if libb2 build needs sse2, how does ports repo have it?
[16:25:34] <cjc7373> Ok, I missed a message, someone already asked
[16:29:51] <bschnei> solskogen would have to check on that. https://github.com 5+ years an issue for it upstream lol
[16:29:52] <phrik> Title: SSE detection is broken on arm64 macos · Issue #36 · BLAKE2/libb2 · GitHub (at github.com)
[16:30:44] <Charon77> I think most of the packages on your list are present on 8.2
[16:31:00] <bschnei> "When --enable-fat is passed the detection for supported compiler flags wrongly enables SSE and AVX"
[16:31:48] <bschnei> Charon77 that's encouraging!
[16:33:17] <Charon77> Big packages that are present in 8.2 from your list: blender, vlc, gstreamer, kodi
[16:35:21] <bschnei> Not surprising. https://gitlab.archlinux.org mentioned media encoding/decoding as a sore spot
[16:35:22] <phrik> Title: Determine minimum required instruction sets (#6) · Issues · Arch Linux / Ports / AArch64 / project-management · GitLab (at gitlab.archlinux.org)
[16:36:38] <Charon77> Could you filter more which ones actually have build errors?
[16:38:00] <bschnei> libb2 is a good case to work on though. What to do when upstream is aware of an issue now for 5 years but can't/won't fix. The issue seems to only affect aarch64 (or non-x64 archs). Do we carry a patch forever? Is that enable-fat option used/necessary? Right now we don't want to remove x64 functionality, but if you were to fix libb2 you'd most likely want to create an MR to patch their configure
[16:38:41] <bschnei> all of those had build errors at *some* point. they may not right now
[16:39:14] <bschnei> the ones towards the top of the list are more likely to have true build errors
[16:39:43] <Charon77> libb2: maybe it's already fixed and that's why it's present on 8.2
[16:39:44] <Charon77> https://github.com
[16:39:45] <phrik> Title: libb2: Fix arm64 compilation by gjasny · Pull Request #64721 · Homebrew/homebrew-core · GitHub (at github.com)
[16:40:14] <bschnei> solskogen may have applied that patch to get it to build and simply not pushed his fork
[16:40:27] <Charon77> oh wait it's not upstream
[16:40:43] <Charon77> so, should we raise an MR to gitlab?
[16:40:47] <bschnei> which is fine, but it still means there is work to do to make the official package actually build for aarch64
[16:40:52] <bschnei> absolutely
[16:42:44] <bschnei> fork libb2, examine that patch from the homebrew project to make sure it's not obviously bad. apply and see if it builds for aarch64. then *make sure it still builds fine for x64*. if you have success in both cases, you're ready to create an MR with a describe commit message as to why the patch is needed
[16:43:59] <cjc7373> based on libb2 repo's configure.ac, I think simply remove enable-fat option under non x86-64 arch is fine
[16:44:30] <Charon77> the patch is just that
[16:45:10] <bschnei> nice. now repeat the exercise for the other 999 missing packages :D
[16:45:42] <bschnei> the good news is a lot are just failing on bad hashes or other trivial things
[16:46:18] <bschnei> if its a package that gets updated regularly a bad hash resolves itself...i don't bother creating a MR
[16:46:56] <bschnei> if it's an old project that doesn't have releases, then I'll usually spend the time
[16:47:04] <Charon77> is your system constantly building?
[16:47:37] <cjc7373> how can a bad hash happen? Sources changed their release?
[16:49:02] <bschnei> Charon77: it will be in order to keep all these packages up to date. I didn't use to have all of these until about 2 weeks ago I just let a script run to try to build as many as possible to help those interested in v8 support.
[16:49:34] <bschnei> it's the same for the v8.2 packages. if you look at the status dashboard, there's nearly always something building
[16:50:37] <bschnei> cjc7373: someone smarter than me should probably answer but it happens a lot that a source from github has its hash change
[16:51:52] <bschnei> I think I see it more when the source is git+ instead of just a straight download of a source tarball, but i've seen it in both cases. sometimes upstream is to blame! they'll re-release with the same tag but different commit lol
[16:59:09] <cjc7373> In go ecosystem, doing a release with the same tag will make it being rejected by the whole world lol (that's how goproxy works)
[17:11:09] <bschnei> Charon77: I can make more "passes" over these missing packages to see which ones will build, but doing so generally will happen after I update existing packages. It doesn't do too much good to add more packages if the ones I have I can't keep up to date
[19:16:57] <solskogen|M> linkmauve: I probably just hacked it out. Don't remember.
[19:18:21] <solskogen|M> It was the early stages - I did a lot of stupid stuff :-)
[20:15:51] -!- TheDcoder has quit [Read error: Connection reset by peer]
[20:16:18] -!- TheDcoder has joined #archlinux-ports
[20:21:28] <solskogen|M> Looks like haskell is back on the menu, boys!
[20:34:36] <bschnei> ^ makes my top 10 list for nerdiest things I've ever heard :D
[20:35:04] <solskogen|M> :D
[20:51:19] <solskogen|M> now it's up to felixonmars to upgrade ghc to the greatest and latest version :-)
[21:26:14] -!- f_ has quit [Read error: Connection reset by peer]
[21:26:23] -!- f_ has joined #archlinux-ports
[22:10:12] -!- titus_livius has joined #archlinux-ports
[22:28:41] -!- Antiz has quit [Quit: The Lounge - https://thelounge.chat]
[22:31:05] -!- Antiz has joined #archlinux-ports
[23:53:34] -!- filmroellchen has quit [Quit: filmroellchen]