#
✕ nick:leio
#asahi-alt · 2026-10-02 18:44
leioI semi-regularly use – and — via compose sequences in own text, AI slop is what steals that from good source texts
#asahi-alt · 2026-09-15 09:58
leioI keep losing that y, maybe that makes sense, they are o marching like a certain grouping in the past or smth
#asahi-alt · 2026-09-15 09:57
leiothe omarchy news post about omarch M
#asahi-alt · 2026-09-15 08:44
leio"If you’ve been working on Linux for Apple hardware, whether that’s kernel drivers, firmware, packaging, installers, or machine learning, we’d love to hear from you. There’s a good chance you’ve solved something we’re still chasing, or vice versa. Get in touch at ... Let’s finish what Asahi started." :D
#asahi-alt · 2026-07-13 18:15
leio(multiple of them, with different USE flag combinations)
#asahi-alt · 2026-07-13 18:14
leiothough all of webkit-gtk, qtwebengine, gcc and firefox have binary arm64 packages in gentoo, afaik
#asahi-alt · 2026-07-13 18:14
leioIt should be on a table, not a lap, while doing such package updates
#asahi-alt · 2026-07-13 11:26
leiobecause I assumed there would be no binary packages/caches for these, but I guess as long as e.g. glibc/mesa isn't patched for asahi, it might have them, just like gentoo
#asahi-alt · 2026-07-13 10:27
leio(I believe nixos would be even more compiling?)
#asahi-alt · 2026-07-13 10:27
leioUmutKzl[m]: You can use binary packages where available
#asahi-alt · 2026-05-21 14:06
leiousually won't have any measurable effect).
#asahi-alt · 2026-05-21 14:06
leioberdanka: It is possible to use "main modules" (?) from binhost as well, provided it has a binary package. Many things with default desktop profile and more do have it, at least the major desktop related packages and a plethora more, with possibility to request for others. You would typically only need to compile some applications that aren't included in binhost, which should go fast, or if you insist on some USE flag customizations that don't have a binary
#asahi-alt · 2026-05-21 14:06
leiopackage (many packages have different USE flag combination binary packages and it can tell what to consider changing on yours to end up using a binary package, some might be those you don't care about). I guess you are implying one would want the main modules compiled for the CPU-specific optimizations, but that's entirely optional and without it it's no worse than binary distribution ones, but yes, there is that possibility to do for what you care about (but
#asahi-alt · 2026-05-21 11:01
leioberdanka: Volpidash[m]: Gentoo has a wide range of binary packages available for typical USE flag combinations for arm64 and you don't have to compile "EVERYTHING" from source.
#asahi-alt · 2026-04-30 16:28
leiothere's also that AGit stuff with codeberg, but I guess you used that already
#asahi-alt · 2026-04-30 16:27
leiowell, still a rank below of becoming an official gentoo dev and doing the pushes yourself ;)
#asahi-alt · 2026-04-30 16:27
leioI would say so, yes
#asahi-alt · 2026-04-30 16:12
leioplease put the details of the change into the commit msg too next time, e.g. the "v6.19.14 based kernel with additional CVE-2026-31431 (copy.fail) and asahi downstream pacthces." bit
#asahi-alt · 2026-04-30 16:11
leiowe don't do merge commits and don't use codeberg as the main git tree, but unlike github, codeberg allows to auto-close things this way, while on github it just showed as closed iirc
#asahi-alt · 2026-04-11 23:34
leiothere's a temp/build.log in there in that case, etc
#asahi-alt · 2026-04-11 22:50
leio(sam is a gentoo dev who would want a report to get it fixed)
#asahi-alt · 2026-04-11 15:39
leioTomus[m]: you can also prep what you want next with asahi-bless (has a -n option to only switch for next boot), asahi-startup-disk as a graphical version; and startup disk in macos (I'm not sure that exposes "only next boot" option)
#asahi-alt · 2026-04-10 22:38
leioI think just discuss it with ztrawhcse, he looks to be in #-arm too
#asahi-alt · 2026-04-10 22:37
leioif it's a generic amd64 and arm64 package
#asahi-alt · 2026-04-10 22:36
leiodon't know who, probably ztrawhcse
#asahi-alt · 2026-04-10 22:36
leiosure, just ask if it's a generic arm64 stable package
#asahi-alt · 2026-04-10 22:34
leioTomus[m]: see also: https://wiki.gentoo.org/wiki/Gentoo_Binary_Host_Quickstart
#asahi-alt · 2026-03-28 08:22
leiois that stuff fatal to the boot process?
#asahi-alt · 2026-03-25 13:17
leiomight it not be waiting on CanGraphical=yes on the seat or something?
#asahi-alt · 2026-02-28 14:39
leio<EliSaado[m]1> "Tomus: I use the repo you..." <- Yeah it’s the exact same for me but if there was a smaller config it could’ve probably taken much less time and the kernel would’ve used less ram as it’d be smaller
#asahi-alt · 2026-02-28 14:39
leioreply looks up mostly OKish on IRC side. It's the edits and such that are annoying; the reply looks like this:
#asahi-alt · 2026-02-26 18:17
leioif not anything else, then because not the channel for this :)
#asahi-alt · 2026-02-26 18:17
leioyeah, I'm going to keep quiet about my nix opinions
#asahi-alt · 2026-02-26 18:15
leiolibreoffice takes 12 minutes here, but there's binary packages to make that 10 seconds or so; I don't see a collabora online package
#asahi-alt · 2026-02-26 18:14
leiook, missed a second l then
#asahi-alt · 2026-02-26 18:13
leioI don't know what colabora is
#asahi-alt · 2026-02-26 18:13
leiothe trick is to not watch the compilation and just do it when you don't care that it goes vroom; or use binary packages where binhost USE flags are fine for it
#asahi-alt · 2026-02-26 18:11
leioyes, it is easy, but I'm a bit biased when I say that
#asahi-alt · 2026-02-17 14:36
leiovirtual/dist-kernel:asahi looks correct to me, but depends on the intent
#asahi-alt · 2026-02-17 14:12
leioJamesCalligeros[m]: I can't read the backlog right now, but you can PM me what exactly needs changing from what to exactly what, and I can sort that out in that case
#asahi-alt · 2026-02-16 07:50
leioI shouldn't have been checking the +i8mm stupidly against my own M2 flags
#asahi-alt · 2026-02-16 06:19
leioah, ok, just additions as you hadn't ran with the cpuid2cpuflags output yet or so
#asahi-alt · 2026-02-16 06:09
leioas I would have expected around 0 rebuilds from that, so might be a good list to review (as a gentoo dev)
#asahi-alt · 2026-02-16 06:09
leiodo you have the list still?
#asahi-alt · 2026-02-16 05:57
leiobut I mostly agree, just would at least at some point clean it up by recompiling what didn't naturally get away from those CFLAGS
#asahi-alt · 2026-02-16 05:55
leioif that doesn't hurt, then why did ffmpeg break
#asahi-alt · 2026-02-16 05:55
leiobut it was armv8.5-a, isn't M1 a armv8.4-a?
#asahi-alt · 2026-02-16 05:52
leioif you just started, most things might be safe from stage3 still
#asahi-alt · 2026-02-16 05:52
leiobut if your -e gets interrupted at some point, might be handy to not rebuild all those things yet again.
#asahi-alt · 2026-02-16 05:51
leiobut that's me, to others I'd have to say -e :D

plus ›