#
✕ nick:sven
#asahi-dev · 2025-03-29 08:31
svenwow, totally missed that work
#asahi-dev · 2025-03-23 15:28
svenbroken MMU can happen with incorrect chicken bits unfortunately
#asahi-dev · 2025-03-23 12:37
svenlol
#asahi-dev · 2025-03-23 11:46
sveniirc the chickens are some of the very first things xnu does anyway
#asahi-dev · 2025-03-23 11:32
svennice :)
#asahi-dev · 2025-03-23 11:17
svenit's pretty likely 0x1f will be somewhere in the xnu binary and mess things up
#asahi-dev · 2025-03-23 11:15
sveniirc there are some control chars that will be caught by the cs software. have you tried disabling them?
#asahi-dev · 2025-03-23 09:12
svenmakes testing and picking them up easier :)
#asahi-dev · 2025-03-23 09:03
svendoes that one also require changes to the dt? if yes please directly add those to the series
#asahi-dev · 2025-03-22 16:43
sventhe chicken setup happens fairly early inside xnu so you don’t even need many thing to work
#asahi-dev · 2025-03-22 16:26
svenrealistically the only way to get the chickens is to grab them from xnu
#asahi-dev · 2025-03-22 12:21
svenor put differently: please briefly check if there's any additional order in that table. if there's nothing obvious i'm happy with "keep the M3 studio below the other M3 machines"
#asahi-dev · 2025-03-22 12:18
svenas i said, i dunno if there's any specific order to the list more than "keep M3s together"
#asahi-dev · 2025-03-22 12:14
svenmight even be sorted by release date or something like that? dunno
#asahi-dev · 2025-03-22 12:13
sveni'd put the M3 device below the other M3s instead of at the end of the list
#asahi-dev · 2025-03-22 12:04
svenhttps://github.com/AsahiLinux/docs/blob/main/docs/hw/devices/device-list.md that should be the source file, you can add them and submit a PR to that repo
#asahi-dev · 2025-03-22 11:22
svenwas starting to doubt my memories of that page
#asahi-dev · 2025-03-22 11:22
svenhah, that’s what I was looking for :)
#asahi-dev · 2025-03-22 11:19
sventhough I guess that doesn’t list the jXXX names
#asahi-dev · 2025-03-22 11:18
sventhere’s also https://asahilinux.org/docs/hw/soc/soc-codenames/
#asahi-dev · 2025-03-22 11:16
svenyeah, same thing usually happens at $work with domain specific language and abbreviations. At the beginning you’re completely lost but after working for a few months you end up speaking in the same code
#asahi-dev · 2025-03-22 10:38
svenand merged :)
#asahi-dev · 2025-03-22 10:37
svenalso, #asahi-offtopic please
#asahi-dev · 2025-03-22 10:35
svenjannau: https://github.com/AsahiLinux/m1n1/pull/453 hope it's fine that i also added you to CODEOWNERS since you know the m1n1 codebase quite well
#asahi-dev · 2025-03-20 21:11
svengetting wifi upstream would also be very nice since that branch should also help some other modern Broadcom boards iirc
#asahi-dev · 2025-03-20 21:11
svenit does
#asahi-dev · 2025-03-19 12:13
svenhrm, I guess I was confused then. I meant to move the constants to the commit that does the chickens. I’ll merge it either way this evening
#asahi-dev · 2025-03-18 22:52
svenor maybe even grab the old commit hash from the PR, I think GitHub shows those before a force push
#asahi-dev · 2025-03-18 22:52
svensure, you can probably and find the commit before the rebase with git reflog
#asahi-dev · 2025-03-18 22:21
sveni'd skip that if you're not confident you have the chicken bits set correctly
#asahi-dev · 2025-03-18 22:20
svenare running with BRINGUP defined? that skips mmu init I think
#asahi-dev · 2025-03-18 22:19
sveniirc only once linux booted though but it was likely already broken inside m1n1
#asahi-dev · 2025-03-18 22:19
svenenabling the MMU without proper chicken bits exploded in the past in weird ways iirc
#asahi-dev · 2025-03-18 22:17
svenin theory chicken bits can effect everything since they're working around cpu errata
#asahi-dev · 2025-03-18 22:17
svenmischa85: i assume you're running with MMU disabled etc?
#asahi-dev · 2025-03-18 22:11
svenonce that's fixed it looks ready to be merged I think
#asahi-dev · 2025-03-18 22:10
svengreat :)
#asahi-dev · 2025-03-18 22:08
svenhttps://github.com/AsahiLinux/m1n1/pull/450/commits/f8d25a3991a5395b075a891e3f02af8be7a3638e that commit should just add two lines or whatever to midr.h
#asahi-dev · 2025-03-18 22:08
sveni'd expect to just see two new lines or so for midr.h
#asahi-dev · 2025-03-18 22:08
svenfl0_id: no, the constants shouldn't be in chickens.c but in midr.h but it looks like they're added to chickens.c in the github diff
#asahi-dev · 2025-03-18 20:31
svenwhich we're supposed to do before picking up any updates to the .dts files in our tree
#asahi-dev · 2025-03-18 20:31
svenit removes an annoying warning whenever we have to run dt_check (or however that target is called)
#asahi-dev · 2025-03-18 20:14
svenfl0_id: okay, CI is fixed but I think you have to rebase on master to fix it in your PR as well
#asahi-dev · 2025-03-18 20:07
svenoh well, let's just merge the fix
#asahi-dev · 2025-03-18 20:01
svendebugging kernel code on a separate mac is less painful and faster than debugging CI issues for me *sigh*
#asahi-dev · 2025-03-18 20:01
svenme neither, but that's common whenever I have to fix something in CI :/
#asahi-dev · 2025-03-18 19:56
svenhah, nice
#asahi-dev · 2025-03-18 19:38
sventried it in a ubuntu vm and it worked there as well
#asahi-dev · 2025-03-18 19:26
svenyeah, i wish i could just ssh into that actions runner and actually debug what's going on
#asahi-dev · 2025-03-18 19:19
svenhate debugging CI issues

plus ›