#asahi-dev · 2026-11-01 22:54
svenwhy?#asahi-dev · 2026-11-01 22:42
nicolas17hm wouldn't it make sense to try making newer macOS / SPTM / exclaves work on M2 or M3?#asahi-dev · 2026-11-01 22:27
svensnooze6214[m]: can you test https://github.com/AsahiLinux/m1n1/pull/758 on the a18 pro? it'll still need that CTRR vreg fix. the exclave stuff is all on the pre-boot volume (and i'm sure also somewhere in the ipsw)#asahi-dev · 2026-11-01 22:18
nicolas17(I rly need to document WKMS too)#asahi-dev · 2026-11-01 22:17
nicolas17or do you just need to derive the key to feed into macOS's aea?#asahi-dev · 2026-11-01 22:16
nicolas17chaos_princess: I assume you found https://theapplewiki.com/wiki/AEA then :P#asahi-dev · 2026-11-01 21:43
jannauI get an iommu read fault and nothing works#asahi-dev · 2026-11-01 21:43
jannauthe 7.3 downstream branch has an issue with https://github.com/torvalds/linux/commit/f0ba5fd51ff23077ce66cb9be79d41a4b7c19c5e#asahi-dev · 2026-11-01 21:42
jannauMichaelReeves[m]: I haven't tested you series without other asahi downstream patches yet. might have been one of the patches of bits/080-wifi that still applied on top. I saw Arend's reply and put testing on my TODO#asahi-dev · 2026-11-01 21:25
svenyeah, the annoying part was that eating was hard for a day or two :D#asahi-dev · 2026-11-01 21:25
fl0_idgood luck, just kinda annoying but should be fine...#asahi-dev · 2026-11-01 21:24
svenouch.. good luck. I only had two of mine removed and the surgery wasn't bad, recovery just sucks.#asahi-dev · 2026-11-01 21:22
MichaelReeves[m]thank you!! I might be inactive for the next bit cause I’m about to have wisdom teeth surgery today (ehh….) but I’ll check it out when I’m up to it#asahi-dev · 2026-11-01 21:20
svenI can test that tomorrow evening#asahi-dev · 2026-11-01 21:20
svenoh.. I missed that#asahi-dev · 2026-11-01 21:20
MichaelReeves[m]yeah the reason I was asking is because jannau said that it didn’t work on m1/m2 for some reason#asahi-dev · 2026-11-01 21:20
sventhat driver is not just used for apple silicon so he's probably hoping people with other machines test as well#asahi-dev · 2026-11-01 21:19
svenyeah, he sent that to the ML as well. sounds like he's happy with your patches and is just waiting a few days for people to test :)#asahi-dev · 2026-11-01 21:12
MichaelReeves[m]ooh the Broadcom maintainer emailed me and apparently they tested it and there weren’t regressions? Not sure what’s going on with m1/m2 if someone could test and give dmesg would appreciate it#asahi-dev · 2026-11-01 13:03
adityaAlr thanks for confirming#asahi-dev · 2026-11-01 13:01
chadmedi dont think we've ever had ipv6 specific issues#asahi-dev · 2026-11-01 13:01
aditya*was#asahi-dev · 2026-11-01 13:01
adityaOr what a bug fix done in Asahi branch?#asahi-dev · 2026-11-01 13:00
adityaIs ipv6 working fine right from starting ?#asahi-dev · 2026-11-01 13:00
aditya@chadmed: I'd#asahi-dev · 2026-11-01 12:59
chadmedid probably also break this up into multiple files in a pcie/ dir at this point it's getting a bit hard to reason about#asahi-dev · 2026-11-01 12:59
adityaMaybe a 4364 specific issue#asahi-dev · 2026-11-01 12:59
chadmedyuka: easiest thing to do would be just branch on the compatible like whats already there, but if we want to make this Nice you should probably do something like a struct of function pointers with model-specific implementations of init/regs/etc maybe#asahi-dev · 2026-11-01 12:59
adityaMaybe a. 4364#asahi-dev · 2026-11-01 12:59
adityaYet to test 4377#asahi-dev · 2026-11-01 12:58
adityaWe have issues with BCM4364#asahi-dev · 2026-11-01 12:56
chadmedaditya: i have not noticed any performance issues with ipv6 on any of my devices#asahi-dev · 2026-11-01 12:56
adityaDoes Asahi also require a patch to disable ARP/ND offload or should it be restricted to T2?#asahi-dev · 2026-11-01 12:55
adityahttps://github.com/t2linux/T2-Debian-and-Ubuntu-Kernel/issues/227#issuecomment-6536502127#asahi-dev · 2026-11-01 12:55
adityaTurning ARP/ND offload off makes speeds better on ipv6#asahi-dev · 2026-11-01 12:50
yuka* in the end#asahi-dev · 2026-11-01 12:50
yukanot sure yet how to do the entrypoint in the en, like whether it makes to preserve one shared top-level init func and then call function pointers from there#asahi-dev · 2026-11-01 12:46
yukaif you have strong opinions, you can leave a comment on the GH commit, this is not far enough to open a PR yet ;)#asahi-dev · 2026-11-01 12:45
yukastill need to split init_port into smaller functions too#asahi-dev · 2026-11-01 12:45
yuka(just an outline, does not compile, but to give you an idea of what I'm planning)#asahi-dev · 2026-11-01 12:44
yukahttps://github.com/yuyuyureka/m1n1/commit/68e5ecf47d3bb86f85b566d88dfda1fe073c7375#asahi-dev · 2026-11-01 12:44
yukastarted splitting up pcie_init_controller() into many small subfunctions which are individually customized#asahi-dev · 2026-11-01 12:35
adityaAre you all also facing an issue where transfer speeds are much slower on ipv6 than ipv4 on wifi?#asahi-dev · 2026-11-01 12:15
yukacan confirm the cert is fixed#asahi-dev · 2026-11-01 12:03
noopwafelshould work now! lmk if you need other traces#asahi-dev · 2026-11-01 11:53
noopwafelat least for t8122 there are a bunch of people who can test :) looking at the cert horror ..#asahi-dev · 2026-11-01 11:49
yukaI know how brittle this whole thing is#asahi-dev · 2026-11-01 11:49
yukathe limitation is not having each and every device to test ;)#asahi-dev · 2026-11-01 11:49
yukayes fair#asahi-dev · 2026-11-01 11:47
noopwafelI think already we're deviating quite a lot from macos trace with that kind of thing