#asahi · 2023-12-23 12:15
kidplayer666Because I do not have enough ram#asahi · 2023-12-23 12:15
kidplayer666How do I do it#asahi · 2023-12-23 12:15
PaulFertserkidplayer666: normally Linux uses ~all free RAM for caching access to block devices because why not. Caches can always be dropped right away when RAM is needed.#asahi · 2023-12-23 12:11
kidplayer666can anyone explain why is there apparently so much on cache instead of on ram??#asahi · 2023-12-23 12:10
kidplayer666https://usercontent.irccloud-cdn.com/file/cQXV7Qux/Screenshot_20231223_113946.png#asahi · 2023-12-23 09:33
pitusti got about one second of audio, then i ^C'ed aplay, and now no more audio happens#asahi · 2023-12-23 08:43
pitustmarcan: "ALSA routes are probably wrong" how should they be configured? also, "aplay output.wav -D 'sysdefault:CARD=AppleJ293'" doesn't produce any audio either (not sure if that's the correct scenario or not)#asahi · 2023-12-22 22:50
Tramtristthanks janneg .. I feel comfortable doing it though need to consider#asahi · 2023-12-22 21:49
leioj`ey: I filed https://bugs.gentoo.org/920555 to start with on the second#asahi · 2023-12-22 21:48
jannegTramtrist: if you manually resize the macos partition to the free up as much space as you want, you can select a much smaller size for the os and use the remaining free space to create your home partition (if you feel comfortable enough to that and linux setup for that manually)#asahi · 2023-12-22 20:48
as400these are not partitions but you still have btrfs quota#asahi · 2023-12-22 20:48
Tramtristoops haha#asahi · 2023-12-22 20:48
Tramtristlsblk#asahi · 2023-12-22 20:48
leioI made a /home after the fact in front of it via letting the installer after initial install free up some space and take from there manually (caring for some partition numbers)#asahi · 2023-12-22 20:47
as400Tramtrist: you get one btrfs subvolume for root and a second one for home#asahi · 2023-12-22 20:47
Tramtristthanks j`ey #asahi · 2023-12-22 20:46
TramtristOn second thought if I go down that road Ill probably go luks and do the volume groups etc#asahi · 2023-12-22 20:46
j`eyTramtrist: yup just one partition for now#asahi · 2023-12-22 20:45
leiogdb has all corrupted stack, it's pointer authentication extension stuff#asahi · 2023-12-22 20:45
j`eyleio: can you see what instruction it fails on?#asahi · 2023-12-22 20:44
Tramtristbeen a while.. does the Asahi installer allow us to create a /home partition manually or is it all or nothing / out of the box?#asahi · 2023-12-22 20:44
leiobytecode_builtins_list_generator and gen-regexp-special-case #asahi · 2023-12-22 20:43
j`eyok that's weird :)#asahi · 2023-12-22 20:42
leioit's a tiny build tool that crashes as part of the build process#asahi · 2023-12-22 20:42
j`eyqtwebengine probably has some hand rolled asm#asahi · 2023-12-22 20:41
leiobut at this point I need to nerdsnipe someone into investigating further *g#asahi · 2023-12-22 20:39
leiono clue how widespread that problem is, I switched to clang from the near start, just now was asked to test qtwebengine with gcc too before ~arm64'ing it#asahi · 2023-12-22 20:37
leiosecond one is that some build tool compiled and used in qtwebengine build just segfaults with +pauth (so once you actually get it to take the -march)#asahi · 2023-12-22 20:37
j`eyit;s funny seeing coworkers names on bug reports :P#asahi · 2023-12-22 20:36
leiohttps://gcc.gnu.org/bugzilla/show_bug.cgi?id=110901 leading to fun like https://bugs.gentoo.org/830165#c11 which reveals that if you don't pass BOTH -march and -mcpu=native in gentoo, or spell out -march=armv8-a+..., you aren't really getting any arch optimizations whatsoever above baseline aarch64#asahi · 2023-12-22 20:34
leiofirst one, sure, a minute#asahi · 2023-12-22 20:33
leionot for the second one, just narrowed it down to the CPU feature flag triggering it#asahi · 2023-12-22 20:32
j`eyleio: dont leave us hanging, got a bug report?#asahi · 2023-12-22 20:32
gfawcett /help#asahi · 2023-12-22 20:23
leioI found my second gcc bug on apple silicon, what do I win? :)#asahi · 2023-12-22 17:46
ydaltonbut in embedded devices the gpu is completely seperate from the display controller#asahi · 2023-12-22 17:46
ydaltonin x86 land, there's basically no distinction between the display controller and the gpu, they are part of the same entity#asahi · 2023-12-22 17:44
ydaltonlovely, x86 monoculture strikes again#asahi · 2023-12-22 17:43
chaos_princessthe answer is probably "x86 does it that way and everything is used to it being done x86 way and will break otherwise"#asahi · 2023-12-22 17:42
marcanit's not useless, you can use it for rendering too#asahi · 2023-12-22 17:42
ydaltoneven if that cardX node is useless otherwise#asahi · 2023-12-22 17:42
ydaltonbruh#asahi · 2023-12-22 17:42
marcanbasically any render-only device still gets a cardX device#asahi · 2023-12-22 17:42
marcanit's complicated(TM)#asahi · 2023-12-22 17:37
ydaltonisn't that what the renderD128 node is for?#asahi · 2023-12-22 17:37
sventhat’s how userspace talks to the kernel driver #asahi · 2023-12-22 17:37
svenit’s a render-only node #asahi · 2023-12-22 17:33
ydaltonon my system, card0 is appledrm and card1 is asahi#asahi · 2023-12-22 17:33
j`ey(check gpu/drm/apple/ for the modesetting part)#asahi · 2023-12-22 17:33
ydaltonthought so, but i don't get why it has a cardX node