#
#asahi-dev · 2025-05-29 21:30
sventhe truly weird part is that sometimes bringing up core 1 works and then it fails at the same place for core 2
#asahi-dev · 2025-05-29 21:30
jannaubetter debugability of the host m1n1 than writing to random registers would be nice
#asahi-dev · 2025-05-29 21:29
jannaumy suspicion are guest intercepts but I'm puzzled why those would only break in this specific place. I also suspect the issue would vanish if llvm would save SP at function entry
#asahi-dev · 2025-05-29 20:38
sven 00125f7c fd 7b bf a9 stp x29,x30,[sp, #local_10]! <-- so that's x29 and lr before it branches to supports_pan. by the time it returns both seem to be corrupted
#asahi-dev · 2025-05-29 20:18
svenjannau: so i'm even more confused by that mmu_setup_secondary abort. i also see a different address inside m1n1 for x29 vs sp which looks reasonable. maybe x29 gets corrupted as well? but i dont see how :/
#asahi-dev · 2025-05-29 20:17
svene.g. the device one just sets the right bit in DWC3_GCTL_PRTCAP
#asahi-dev · 2025-05-29 20:16
svenon m1+ the dwc3 tunables in the ADT are just documented writes to the normal dwc3 regs iirc
#asahi-dev · 2025-05-29 19:42
amarioguyah
#asahi-dev · 2025-05-29 19:29
nickchanin particular cfg0-device cfg1-device cfg0-host cfg1-host properties from the adt
#asahi-dev · 2025-05-29 19:28
nickchanthey aren't device specific but i think they could be specific down to a SoC revision
#asahi-dev · 2025-05-29 19:28
nickchanamarioguy: am just meaning i could use the tunables logic for somw usb2 phy tunables
#asahi-dev · 2025-05-29 18:36
amarioguywell t2/a10x
#asahi-dev · 2025-05-29 18:36
amarioguynickchan: wait why would t8015 require usb3? or is this for t2?
#asahi-dev · 2025-05-29 16:34
kettenismy OpenBSD-built chainloading m1n1 binaries work, so I guess that removes the issue with rust in m1n1 from my side
#asahi-dev · 2025-05-29 15:23
sven*them
#asahi-dev · 2025-05-29 15:23
svenatcphy needs then because what Apple calls „tunables“ are there is in reality 10% tunables and 90% init sequence they didn’t want to put into the atcphy kext for whatever reason
#asahi-dev · 2025-05-29 15:20
nickchanand by tunables I mean device specific tunables
#asahi-dev · 2025-05-29 15:19
nickchani think usb3 is not sensitive enough to absolutely require tunables do they
#asahi-dev · 2025-05-29 15:18
nickchanyay! I need similar logic too
#asahi-dev · 2025-05-29 14:47
alyssayay!
#asahi-dev · 2025-05-29 14:24
svenhttps://git.kernel.org/pub/scm/linux/kernel/git/sven/linux.git/log/?h=wip/atcphy and https://github.com/AsahiLinux/m1n1/pull/469 handles fuses as if they were another tunable and I think I like it that way
#asahi-dev · 2025-05-29 10:15
svenlol
#asahi-dev · 2025-05-29 10:15
chadmed:)
#asahi-dev · 2025-05-29 10:14
mazcheers! :)
#asahi-dev · 2025-05-29 10:14
chadmed(as in yes go ahead, no i dont mind)
#asahi-dev · 2025-05-29 10:14
chadmedhaha yeah sure
#asahi-dev · 2025-05-29 10:13
mazchadmed: I like the analogy! mind if I quote you? :)
#asahi-dev · 2025-05-29 10:12
chadmedmaz: carburetted interrupt controller? /s
#asahi-dev · 2025-05-29 07:53
mazand since AIC2 is pretty retarded on that front (no direct injection), it's not like we're losing much.
#asahi-dev · 2025-05-29 07:52
mazyou "just" leave some performance on the table if you can't control it.
#asahi-dev · 2025-05-29 07:52
mazKVM doesn't care about the routing of the interrupts as long as they are purely virtual
#asahi-dev · 2025-05-29 07:52
mazKVm doesn'
#asahi-dev · 2025-05-29 07:51
mazamarioguy: I'm not quite dead yet! :)
#asahi-dev · 2025-05-29 01:28
amarioguymaz: not sure if you're still active so apologies in advance for the ping, but does kvm exactly care much about AIC2 seemingly routes interrupts to cores at random? didn't seem like there was any special code to do that in either the AIC driver or the KVM vgic code, but then again the fact that virtual CPU state is entirely maintained by the HV means that it may not matter much i suppose
#asahi-dev · 2025-05-29 01:05
amarioguy(ik you can have CPUs opt out of interrupts via the SIQ cfg msr but iirc that only allows filtering between P and E core clusters
#asahi-dev · 2025-05-29 01:04
amarioguyjust refreshing my memory here since i'm at the point where i have to finish the vgic implementation now: AICv2 and v3 i hear don't support setting CPU affinities?
#asahi-dev · 2025-05-28 14:33
fl0_id(I quoted those from memory so they might not be exactly correct)
#asahi-dev · 2025-05-28 14:32
fl0_idzink_context, zink_screen and memory page sizes for vn_renderer seemed most likely to be relevant
#asahi-dev · 2025-05-28 14:31
fl0_idchaos_princess well those were the patches being applied in the fedora package. some seemed unrelated like "gnome-sth-fix" and others obsolete. But you could very well be right ofc :)
#asahi-dev · 2025-05-28 13:57
chaos_princessi am not sure if the part you are looking at is relevant for macos either
#asahi-dev · 2025-05-28 13:56
fl0_idyeah, I assume so. but I have no idea which parts might only be relevant for one or the other :)
#asahi-dev · 2025-05-28 13:56
chaos_princessit is the same(?) similar(?) technology between podman desktop and muvm
#asahi-dev · 2025-05-28 13:54
fl0_idthx
#asahi-dev · 2025-05-28 13:54
fl0_idsry for the OT and confusion
#asahi-dev · 2025-05-28 13:54
chaos_princesshttps://app.element.io/#/room/#asahi:fedoraproject.org - the person you are looking for is Sergio Lopez
#asahi-dev · 2025-05-28 13:53
fl0_idpodman, with or without desktop but on macos. that's why I only asked if this person was here sometimes originally...
#asahi-dev · 2025-05-28 13:52
chaos_princessare we talking containers in linux sense, or "podman desktop on macos" sense?
#asahi-dev · 2025-05-28 13:48
fl0_idthis it at least also still the state described in podman documentation
#asahi-dev · 2025-05-28 13:48
fl0_idchadmed not for asahi, but afaik it is needed for allowing container access to the gpu (not the raw gpu, but still) I know it is OT, which is why I originally only asked if someone knew this person, as he also seemed to be involved with asahi based on their blog
#asahi-dev · 2025-05-28 13:44
chadmedwhat for though? its not needed for anything afaict

plus ›