#
…

2026-01-17

00:01alxdrl has quit [Quit: Textual IRC Client: www.textualapp.com]
00:02hdbngr has quit [Ping timeout: 480 seconds]
00:10bigsilicon has joined #asahi-dev
00:17maor26 has quit [Ping timeout: 480 seconds]
00:18bigsilicon has quit [Ping timeout: 480 seconds]
00:26cremfuelled has quit [Quit: cremfuelled]
00:26michaelrostom has quit [Quit: Konversation terminated!]
00:32hdbngr has joined #asahi-dev
00:37michaelrostom has joined #asahi-dev
00:43fl0_id___ has joined #asahi-dev
00:46yukaI debugged my minimal device tree with a single cpu core on t8132 baremetal further. Copied the debug_putc into the kernel and added debug prints in the very first code of the kernel.
00:47yukaAs a reminder, I was not seeing any output after "Vectoring to next stage" despite having the serial console enabled in the device tree and setting earlycon
00:47yukaWith the help of debug_putc, I traced it into the __enable_mmu function in head.S of the kernel
00:49yuka(Or maybe enabling the mmu is preventing debug_putc from working afterwards)
00:50hdbngr has quit [Ping timeout: 480 seconds]
00:55fl0_id___yuka, this seems very similar to what i am seeing
00:55yukaJust getting no output when starting the kernel?
00:56fl0_id___though I didn't trace it to head.S but build_all_zonelists
00:56fl0_id___yes, after vectoring to next stage
00:56fl0_id___at least I have symbols now after disabling random base
00:58yukaIt seems CONFIG_DEBUG_LL would be useful for getting even earlier uart
01:01mixi has quit [Ping timeout: 480 seconds]
01:04pb17 has quit [Ping timeout: 480 seconds]
01:04pb17 has joined #asahi-dev
01:06hdbngr has joined #asahi-dev
01:07fl0_id___yuka, it seems like this is not available for arm64, no?
01:07yukaIs this stacktrace from earlycon output?
01:08yukaThere is code related to it in the arm64 head.S
01:09yukaSo possibly not available out of the box, but it should be possible to be inspired by this code to make my own debug_putc work after mmu init
01:10fl0_id___yuka, no this stacktrace from hv over m1n1 usb
01:11fl0_id___I should prob use the cs m1n1 connection though
01:14hdbngr has quit [Ping timeout: 480 seconds]
01:14mixi has joined #asahi-dev
01:17michaelrostom has quit [Quit: Konversation terminated!]
01:33hdbngr has joined #asahi-dev
01:38fl0_id___there should at least be pr_notice("Kernel command line: %s\n", saved_command_line); before the problem section
01:42hdbngr has quit [Ping timeout: 480 seconds]
01:47fl0_id___and I also put anohter printk which is also not shown
02:27hdbngr has joined #asahi-dev
02:35hdbngr has quit [Ping timeout: 480 seconds]
03:03tobhe has joined #asahi-dev
03:10pb17 has quit [Ping timeout: 480 seconds]
03:10tobhe_ has quit [Ping timeout: 480 seconds]
03:14pb17 has joined #asahi-dev
03:16ralim has quit [Quit: The Lounge - https://thelounge.chat]
03:16hexdump01 has joined #asahi-dev
03:16ralim has joined #asahi-dev
03:18hexdump0815 has quit [Ping timeout: 480 seconds]
03:55hdbngr has joined #asahi-dev
04:04hdbngr has quit [Ping timeout: 480 seconds]
04:14integralpilot has joined #asahi-dev
04:18nora has joined #asahi-dev
04:20nora_ has quit [Ping timeout: 480 seconds]
04:23hdbngr has joined #asahi-dev
04:31hdbngr has quit [Ping timeout: 480 seconds]
04:56Yushi has joined #asahi-dev
05:11integralpilotfl0_id: given you had config issues before, can you confirm that CONFIG_APPLE_AIC is set to y in your .config?
05:12integralpilotShiz: omg yay that's so exicting, great work!
05:12ansg191 has quit [Ping timeout: 480 seconds]
05:12Shizmuch thanks to your earlier efforts too :)
05:12integralpilothow'd you solve the memory corruption, i was quite confused & stuck about it. was it something with limiting to just one core, i thought i remember something about secondary CPU startup being correlated but not sure.
05:13ansg191 has joined #asahi-dev
05:14ShizI think it was a combination of three things:
05:15Shiz1) the PMGR power-gating hacks were definitely poking addresses they shouldn't be (and entirely differently mapped on M3+)
05:15ansg191_ has joined #asahi-dev
05:15Shiz2) adding some MSRs and marking some more memory as reserved using the memory-map ADT node
05:16Shiz3) booting with only one P-core and only one E-core enabled
05:16ansg191 has quit [Remote host closed the connection]
05:16Shizomitting any of those seems to make it return to unstable boot, so there's def still work to be done :p (mostly the last one)
05:16integralpilotoh interesting, great work on finding all those! :)
05:16Shizit seems `self.add_tracer(zone, "CPU_START", TraceMode.RESERVED)` is also messed up, needs to be commented-out
05:16ansg191_ is now known as ansg191
05:16Shizmessy but at least a good base for further research :D
05:19Shizat least half of it seems to come down to "whoops the m3 has a very different memory/mmio map and maybe m1n1 was making a few assumptions related to how m1/m2 were mapping"
05:19Shizdirectly or indirectly
05:20Shiz(no shade of course, you can never foresee the entire future)
05:26hdbngr has joined #asahi-dev
05:33JamesCalligeros[m]Yeah like lmfao in the faq under “wont apple just change everything?” the counter example i gave was AIC, which they immediately changed and then changed again
05:33JamesCalligeros[m]So yes, some incorrect assumptions were made early on
05:34NickChan[m]the things mostly don't change assumption is still true
05:34NickChan[m]But mac is the moment they have to make big chips
05:34NickChan[m]So a bunch of things changed
05:35JamesCalligeros[m]Yeah, the core stuff doesnt really change much architecturally
05:35JamesCalligeros[m]They do like moving stupid crap around for some reason, like epic endpoints for dcp methods
05:36chadmedyo dawg i heard you like unstable interfaces so we put an unstable interface inside your unstable interface
05:37hdbngr has quit [Ping timeout: 480 seconds]
05:40lion328 has quit [Ping timeout: 480 seconds]
05:47hdbngr has joined #asahi-dev
05:56hdbngr has quit [Ping timeout: 480 seconds]
05:58rosalyntg has quit [Remote host closed the connection]
06:03rosalyntg has joined #asahi-dev
06:09__Tetris__ has quit [Ping timeout: 480 seconds]
06:11rosalyntg has quit [Remote host closed the connection]
06:13hdbngr has joined #asahi-dev
06:22hdbngr has quit [Ping timeout: 480 seconds]
06:30rosalyntg has joined #asahi-dev
06:36__Tetris__ has joined #asahi-dev
06:44__Tetris__8 has joined #asahi-dev
06:45__Tetris__ has quit [Ping timeout: 480 seconds]
06:50__Tetris__ has joined #asahi-dev
06:53__Tetris__8 has quit [Ping timeout: 480 seconds]
07:02__Tetris__ has quit [Ping timeout: 480 seconds]
07:11ouethy has joined #asahi-dev
07:16ACG_ has quit [Remote host closed the connection]
07:16hdbngr has joined #asahi-dev
07:26maor26 has joined #asahi-dev
07:29hdbngr has quit [Ping timeout: 480 seconds]
07:30hdbngr has joined #asahi-dev
07:31__Tetris__ has joined #asahi-dev
07:39hdbngr has quit [Ping timeout: 480 seconds]
07:52hdbngr has joined #asahi-dev
08:00weirdtreething has quit [Remote host closed the connection]
08:00weirdtreething has joined #asahi-dev
08:01hdbngr has quit [Ping timeout: 480 seconds]
08:07Bjarne[m] has joined #asahi-dev
08:09LarstiQnoopwafel, Shiz: very nice work!
08:11Yushi has quit [Remote host closed the connection]
08:31hdbngr has joined #asahi-dev
08:39hdbngr has quit [Ping timeout: 480 seconds]
08:44Yushi has joined #asahi-dev
08:45svenso you’re booting with two cores and that works but any more and it’s unstable again? Weird
08:51integralpilotjust pushed new MTP/dockchanel stuff after applying sven's feedback and making some changes. thanks again for making the branch jannau, i think the PR UI is much easier for this!
08:52svennice, will take another look either this afternoon or tomorrow probably
08:53riker77 has quit [Quit: Quitting IRC - gone for good...]
08:54noopwafelintegralpilot: this is all on top of your notes (for the bringup) of course, thank you a lot for writing/sharing..
08:55nickchanmaybe check if it works with 1 core per cluster
08:55nickchanneeds the bigger chips obviously
08:58midou has quit [Ping timeout: 480 seconds]
08:59svenit’s gonna suck if this some long standing locking or whatever issue that we only hit now because m3 speculates even harder than m1/2 or something equally cursed 🙈
09:01noopwafelwhat is the status on the linux side? I have not tried booting a kernel yet. nvme?
09:02Yushi has quit [Remote host closed the connection]
09:03nickchanas much as i don't like to say this but
09:03nickchantry reverting the lock changes in src/utils.c
09:05Yushi has joined #asahi-dev
09:05nickchaneven though i think the logic there is sound
09:07integralpilot has quit [Ping timeout: 480 seconds]
09:08riker77 has joined #asahi-dev
09:10svennoopwafel: afaik there are issues with nvme, integralpilot has managed to boot a linux kernel to userspace but other people ran into issues so far
09:11svennvme probably just needs someone with the hv to run run_guest.py with -m trace_nvme.py (maybe set verbose=3 in there) and see what changed
09:13noopwafelah cool. I just want to boot to somewhere I can make friends with gpu, so that sounds super hopeful..
09:14jannaunoopwafel: I think nvme should work with 14.x. the breaking isssue is 15.x
09:20noopwafelnickchan: good news, reverting does not fix. I'm not sure it's locking, the boot failures are not _that_ random, although maybe this is just wishful thinking :)
09:22noopwafelmy hope is that continuing down the path of 'working out what everything in the proxy code is actually meant to do' will help.. but probably good to push changes first
09:22nickchanin regards to speculation there's always the giga hack fix which is to enable TSO
09:22noopwafeloh this is a genius idea
09:23svenyup :D
09:26hdbngr has joined #asahi-dev
09:40hdbngr has quit [Ping timeout: 480 seconds]
09:42integralpilot has joined #asahi-dev
09:47noopwafelso none of this helps and also I hit exactly the same failure 10 times in a row
09:58midou has joined #asahi-dev
09:59integralpilotat least its semi-deterministic :D
10:00svenit' just weird that it really doesn't like multiple cpus to be active
10:01integralpilotand mulitple CPUs works totally fine on hypervised linux on M3
10:06nickchanintegralpilot: have you tested what happens if a EL1 guest performs SError-generating accesses on the vGIC
10:07nickchanon m1 that would would generate an SError... to EL2
10:07integralpilotwhat is vGIC, sorry?
10:08nickchanvirtual generic interrupt controller and GIC is the standard arm interrupt controller
10:08nickchanm1/a14+ does have vGIC without the GIC
10:08integralpilotoh, that's interesting!
10:09nickchanon m1 there's an issue if the guest write something that isn't right into the vGIC, it sends an Serror to the host
10:09nickchanWhich would of course crash the host were it not for a workaround in kvm
10:09nickchanBut supposedly that's fixed in m3
10:10hdbngr has joined #asahi-dev
10:10integralpilotoh cool, hopefully it is.
10:10integralpilothow do i make a SError generating access?
10:11integralpilotsorry, don't know much about vGIC / arm interrupts in general.
10:13nickchanokay it's not exactly an errata it's just a specified feature that absolutely nobody else implemented and also deprecated
10:14nickchanand was very under-specified
10:18hdbngr has quit [Ping timeout: 480 seconds]
10:19integralpilotoh that's quite strange
10:20nickchanmaybe on m3 the serror is delivered to the guest or something
10:20nickchan but given that there's 10 lines in the spec there's no asserting which behaivor is correct even though delivering to guest makes more sense
10:21integralpiloti've been doing research and think i understand more now, should be able to test if fixed in M3 when i've got free time. unfortunetly am quite busy for the next few days, if anyone else with M3 wants to test it before me feel free.
10:22chadmed has quit [Read error: Connection reset by peer]
10:23nickchanand they did find another systems that implemented the under-specified feature in a way that makes sense
10:23nickchanso it was changed to a list of midrs
10:23chadmed has joined #asahi-dev
10:26integralpiloti wonder why apple felt the need to even implement this feature?
10:26hdbngr has joined #asahi-dev
10:27midou has quit [Ping timeout: 480 seconds]
10:27nickchanintegralpilot: do you have linux boot logs?
10:27integralpilotas in, dmesg output?
10:27nickchankvm has some logs
10:28integralpilotcan grab it quickly, don't think i've saved that.
10:29nickchani figured it wouldn't help
10:29nickchangiven the changes to kvm over time
10:30integralpilothaven't got KVM enabled so need to rebuild, might take a while (very old hardware for my x86 kernel buildbox), sorry
10:32midou has joined #asahi-dev
10:37integralpilotor do you not think it's worth it?
10:39midou has quit [Read error: No route to host]
10:40svenmight be fun to know, but not very important imho
10:41hdbngr has quit [Ping timeout: 480 seconds]
10:42nickchanit's something that needs to be figured out before release but not neccessarily now
10:43svenwhy before release?
10:43nickchanwell, a broken SEI generation can take down the whole system
10:46hdbngr has joined #asahi-dev
10:47svenoh, because it's a MIDR list and not just enabled on any apple machine
10:47svenbut still... much more important things to do before that
10:47nickchanright and testing this stuff would be easier when there's like a working desktop so not now
10:47svenfirst step would be to get some minimal M3 device tree on LKML
10:48nickchanthe step before that would be upstream gicv3
10:48nickchanerm, aicv3
10:49svensure, that's part of that
10:49svenand the current thing at least needs the generic compatible removed before that's possible
10:49integralpilotupstream aicv3 would actually be useful because currently it's one of the patches i need to apply ontop of upstream whenever testing something i'm working on
10:50integralpilotcan i be submitted on its own or would it need to be in a patchset with adding M3 DTs so it's not dead code?
10:50integralpilot*it be submitted
10:50svenand if we remove the generic compatible anyway we can also hardcode the 3(?) new fdt properties i guess
10:50svenit can be submitted on its own
10:50nickchani don't think DTs are ever a requirement
10:50svenneeds dt-bindings still I think
10:51integralpilotiirc AIC3 code is jannau's
10:51svenwhat i'd do is look at all M3 ADT and collect config-offset,cap0-offset and maxnumirq-offset and see if and how they differ
10:52nickchanmaybe m4 and m5
10:52sventhen come up with dt-bindings based on that which just use "apple,tXXXX-aic3" or something like that
10:52svennickchan: and they need different compatibles anyway, so at least right now I don't see a reason for those apple,config-offset, etc. properties in the FDT
10:53svenstill a good idea to look at m4 and m5 adts as well
10:53svenmaybe there's aic4 and aic5 now :P
10:57hdbngr has quit [Ping timeout: 480 seconds]
11:05integralpiloton M3 and M3 pro: config-offset: 0x10000, cap0-offset: 0x4, maxnumirq-offset: 0xC
11:05integralpilot(same on both)
11:05hdbngr has joined #asahi-dev
11:06integralpilotfl0_id: can you check your M3 max, if it's okay and when you have time?
11:07integralpilothopefully it's consistent across all
11:07svenif it is i'd just use the base M3 compatible as the fallback for all of these and hardcode the offsets in the driver
11:07svenand then just remove those device tree properties
11:07integralpilotwait maybe they reason fl0_id is having trouble is because it's not
11:08integralpilotin the minimal DT i wrote for them i assumed it was
11:10sveneven if it's different i'd just hardcode the values for M3 max and just not use the fallback M3 base compatible there then
11:11integralpilotthat sounds like a good stratergy, thanks!
11:11integralpiloti am downloading t6031 ipsw right now to check
11:12svenlooking at the dt-bindings if we do it like that it just needs the compatibles added to apple,aic2 (though it technically isn't aic2 anymore then :D)
11:12svenand the description adjusted i guess
11:13integralpilotand marcan as maintainer changed to you?
11:13integralpilotor fine to leave as is if touching aic2 yaml?
11:14sveneuh.. i guess we could change that while we're at it
11:14svenall that really matters is that the emails get sent to that asahi list anyway, and that happens due to MAINTAINERS
11:14hdbngr has quit [Ping timeout: 480 seconds]
11:14midou has joined #asahi-dev
11:14integralpilot1 hour and 15 minutes remaining on IPSW download :D
11:15svenouch :D
11:15integralpilotlove my terrible 3MB/s capped internet...
11:15chaos_princessasahi-installer has a py script to download individual files out of zips
11:16integralpilotwell, technically, 3.125MB/s (it's a 25 Mbps plan)
11:16nickchandoes uboot just ignore framebuffer stride? tried to boot uboot on a8 and the screen looks like stride is just ignored
11:17integralpilotchaos_princess: that is really helpful, thank you!
11:19nickchanugh, ./drivers/video/simplefb.c in u-boot makes no mention of stride at all
11:20integralpilotthat's concerning, how'd they get away with that for so long?
11:21integralpilotactually, it was fixed 3 weeks ago :D
11:21integralpilot"video: simplefb: Add stride handling"
11:22nickchantime to cherrypick that
11:30nickchanyay! booted to crash stacktrace
11:31midou has quit [Read error: No route to host]
11:31nickchanright, i just need to fixup some m1-only sysreg access
11:34hdbngr has joined #asahi-dev
11:43hdbngr has quit [Ping timeout: 480 seconds]
11:47fl0_id_ has joined #asahi-dev
11:49fl0_id_integralpilot thx for thinking about that. I am traveling today, so can only check tonight cet
11:50integralpiloti am checking right now :)
11:50fl0_id_ has quit [Remote host closed the connection]
11:51integralpilotthough it is ~11pm aest so might need to log off before i can finish
11:58integralpilotokay so checked, and it's the same on m3 max
11:58lion328 has joined #asahi-dev
11:59integralpilotonly differences on ADT AIC node between pro vs max are the strides, capability mask, reg (ofc), and cpu count, as expected
12:00nickchanoh it's actually crc32 instructions not supported
12:00nickchanguess there's no runtime check
12:02MikeM64 has quit [Read error: Connection reset by peer]
12:02MikeM64 has joined #asahi-dev
12:04integralpilotOK yes, just checked M5 and it STILL uses aic,3
12:04integralpilotso they haven't changed it AGAIN, thankfully
12:05integralpilotsame config-offset (they call it extint-baseaddress), cap0-offset and maxnumirq-offset as M3, m3 Pro and m3 max too
12:05svengreat, so just using the m3 compatible as baseline and hardcoded the values sounds viable then
12:06nickchanand uboot's usage of the wdt expire now don't work on a8, so I will change it to m1n1's wdt_reboot() behavior which works
12:07integralpilotgreat uboot work! :)
12:07integralpilotbtw, anyone curious / want me to check anything re. M5 now that i've got the DT open?
12:10integralpilotin good news, M5 also hasn't changed ascwrap or RTBuddy versions
12:11integralpilotokay doing AIC3 stuff now, are we sure it's okay to bolt onto the apple,aic2 dt binding or is it safer to make a new one?
12:11midou has joined #asahi-dev
12:15integralpilothave to log off now but can look more into it tomorrow! thanks for your help sven working all this out :)
12:15integralpilot has quit [Quit: integralpilot]
12:26yukaHow do you use the m1n1 proxyclient on the central scrutinizer serial?
12:27yukaFor me, I can open the ttyACM device in screen/picocom and see that it's running the proxy, but trying to run the proxyclient on it yields this: https://termbin.com/adby
12:29MikeM64 has quit [Read error: Connection reset by peer]
12:29MikeM64 has joined #asahi-dev
12:31nickchanhmm uboot doesn't work on a9-a11
12:32midou has quit [Remote host closed the connection]
12:32nickchanof course uboot assumes 4K everywhere
12:32nickchana mircale that works on m1
12:32nickchanbut it will not work on hardware that actually does not have 4K paging
12:35hdbngr has joined #asahi-dev
12:36nickchanand uboot has dart translation support so that's working with 4K page on the CPU MMU and 16K IOMMU
12:39jannauyuka: it does not. the interactive control and printing breaks it. someone wanted to rewrite cs firmware so that it offers a transparent serial as well but I don't remember why and haven't heard of it again
12:40yukathanks for confirming
12:43Shizoh, I thought that was what the micro-usb on the actual CS board was for :x
12:43yukait does do serial passthrough
12:43yukaI can see the serial output from m1n1
12:44hdbngr has quit [Ping timeout: 480 seconds]
12:44yukabut the proxy protocol does not work over it because it also has the central scrutinizer control stuff mixed in
12:47kettenisnickchan: there is no fundamental problem with using the DART with 4k pages
12:48kettenisit's just the Linux abstraction layers that make it difficult
12:51midou has joined #asahi-dev
12:58okb has joined #asahi-dev
13:03___nick___ has joined #asahi-dev
13:04midou has quit [Ping timeout: 480 seconds]
13:05svenintegralpilot: i'd put it into the aic2 binding since there's no change except for the compatible
13:06svenyeah, the 4k cpu / 16k iommu thing is just hard due to linux abstractions
13:07sveni had a very old patchset that fixed that but dropped it because it didn't work for the gpu and we also had stuff like muvm at that point
13:12hdbngr has joined #asahi-dev
13:12jannauthe main issue is that you can't map disjoint 4k pages continuously. 4k aligned start/end is supported via the unused sub page protection flags. devices/iommu's would need a flag only use 16k folios for me
13:12midou has joined #asahi-dev
13:21sveniirc there's no iommu api that guarantees continous maps
13:21svenit's just that at least some subsystem (v4l? something like that) abuses the scatter-gatherlist map api which implicitly does that
13:21midou has quit [Read error: No route to host]
13:22___nick___ has quit [Ping timeout: 480 seconds]
13:23sven"With a sufficient number of users of an API, it does not matter what you promise in the contract: all observable behaviours of your system will be depended on by somebody. " ;)
13:26hdbngr has quit [Ping timeout: 480 seconds]
13:33midou has joined #asahi-dev
13:51ouethy has quit [Quit: Konversation terminated!]
13:57ouethy has joined #asahi-dev
14:02ouethy has quit []
14:02ouethy has joined #asahi-dev
14:02ouethy has quit []
14:08kettenisnote that u-boot maps memory 1:1, so there so mapping disjoint 4k pages isn't a big issue
14:14midou has quit [Read error: No route to host]
14:16okb has quit [Ping timeout: 480 seconds]
14:25___nick___ has joined #asahi-dev
14:29hdbngr has joined #asahi-dev
14:31Shiz nickchan : maybe check if it works with 1 core per cluster
14:31Shizunless I misunderstood, that is what we did
14:31Shiz(one P-core, one E-core)
14:33midou has joined #asahi-dev
14:34nickchanShiz: was thinking about bigger chips that has 3 clusters or more
14:34Shizahh ;p
14:35Shizm3 max/ultra i guess
14:35ShizI found myself diving into weird apple hardware for a bit and found out about something called cloudOS which they run on their private compute nodes...
14:35Shizwhich of course also have board IDs :)
14:36okb has joined #asahi-dev
14:36chaos_princessm3pro is single cluster?
14:36chaos_princess*single p-cluster
14:36ShizI wanna say yes, but I'll verify tomorrow :P
14:36Shiz(xnu will crash early if not at least one core in all clusters is activated, so easy to check)
14:37Shizactually, can just check the devtree -- yep, two clusters
14:38Shizcluster 0 is E, cluster 1 is P
14:40pb17 has quit [Ping timeout: 480 seconds]
14:40Shizm3 max does have 3 clusters
14:41nickchanturns out it isn't too hard to add 16K and 64K paging to uboot
14:43kettenisbut why? did apple really make armv8 CPU cores that don't support the 4k page granule?
14:45nickchan4k paging isn't support from a9-a13
14:45nickchanthen in a14 4k paging is added again
14:45nickchanand yes this means DTK's rosetta is cursed
14:47chaos_princessis there kmutil on dtks?
14:47nickchanworks on 11.2
14:49hdbngr has quit [Ping timeout: 480 seconds]
15:05pb17 has joined #asahi-dev
15:21hdbngr has joined #asahi-dev
15:30hdbngr has quit [Ping timeout: 480 seconds]
15:33__nick__ has joined #asahi-dev
15:34midou has quit [Ping timeout: 480 seconds]
15:36___nick___ has quit [Ping timeout: 480 seconds]
15:37hdbngr has joined #asahi-dev
15:43nickchanokay i will use 64k paging in uboot so i can ship only 1 bin
15:59bigsilicon has joined #asahi-dev
16:02midou has joined #asahi-dev
16:07bigsilicon has quit [Quit: bigsilicon]
16:08bigsilicon has joined #asahi-dev
16:10hdbngr has quit [Ping timeout: 480 seconds]
16:17midou has quit [Remote host closed the connection]
16:18bigsilicon has quit [Ping timeout: 480 seconds]
16:25bigsilicon has joined #asahi-dev
16:26midou has joined #asahi-dev
16:28svenhm... i fixed the weird assert but now nothing shows up in boltctl anymore
16:29svenand i'm not sure if stuff showed up when it still triggered that assert
16:30hdbngr has joined #asahi-dev
16:36bigsilicon has quit [Ping timeout: 480 seconds]
16:40jannauoh, usb 1.1/2.0 devices sometimes survived suspend/resume. I wasn't aware and I replugged my yubikey always over suspend/resume
16:42midou has quit [Ping timeout: 480 seconds]
16:44aissen_ has quit [Ping timeout: 480 seconds]
16:48midou has joined #asahi-dev
16:49svenhuh, I thought they all died
16:52jannaume too but a logitech unifying receiver dongle survives on both 6.17/6.18, the usb 2.0 hub of my usb3/dp-alt hub survives on 6.17 but not 6.18
16:52okb has quit [Ping timeout: 480 seconds]
16:53jannaubehaviour on 6.18 might be explained by the additional always on power-domain for the DP enabled port
16:54okb has joined #asahi-dev
16:56jannauno, unifying receiver survived on the non-DP port as well
16:56aissen has joined #asahi-dev
17:00svenhrm, so either I messed something or something in drivers/thunderbolt changed between early 2025 or so and now :(
17:02pb17 has quit [Ping timeout: 480 seconds]
17:03aissen_ has joined #asahi-dev
17:09jannaunot that many changes in v6.12..v6.18 ignoring new intel platforms documentation and debugfs changes
17:09aissen has quit [Ping timeout: 480 seconds]
17:10WindowPain has quit [Quit: Ping timeout (120 seconds)]
17:10WindowPain7 has joined #asahi-dev
17:10WindowPain7 is now known as WindowPain
17:10svenyeah, don’t see anything suspicious either
17:12svenright now it finds the host router and establishes the domain but then it doesn’t see the connected device after I set that Apple specific VSE reg in the root switch
17:12svenI suspect I messed up somewhere around there
17:33aissen has joined #asahi-dev
17:33aissen_ has quit [Read error: Connection reset by peer]
17:46hdbngr has quit [Ping timeout: 480 seconds]
17:53pb17 has joined #asahi-dev
17:58okb has quit [Read error: Connection reset by peer]
18:14bigsilicon has joined #asahi-dev
18:25bigsilicon has quit [Quit: bigsilicon]
18:25bigsilicon has joined #asahi-dev
18:26jannausigh, did apple break installing Sonoma on M2 devices? usb install media installer fails aftyer a while with "failed to download insteller asset (SFR)" and ask to check the internet connection. once with wifi, once with lan
18:27Shizhave you tried using Mist?
18:28Shizor derp, is this from when you're already booted into the usb
18:30jannauthis is during the install. creating the install media works fine and the same usb stick worked on m3. the installer writes quite a bit onto the volume
18:31jannauI'm installing ventura now and see if I can upgrade
18:37chaos_princesswhich sonoma have you tried?
18:48LarstiQsven: ah Hyrum's law. Makes me wish that I knew exactly who the users of our apis are sometimes
18:48fl0_id___integralpilot yes CONFIG_APPLE_AIC=y
18:49Bjarne[m] crate as Hey, I wanted to contribute something and found the open "Rewrite tuxvdmtool to userspace i2c" Task on your Help Wanted page. I rewrote the cd321x driver logic in rust using the i2cdev crate as proposed.
18:49Bjarne[m]In the [issue](https://github.com/AsahiLinux/tuxvdmtool/issues/1) there is also the point of building a single tools with multiple backends. How should this look like?
18:49Bjarne[m]s/crate as//
18:49jannau14.8.3, ventura 13.5 failed as well, I'll update the main os from 15.0
18:49Bjarne[m]* Hey, I wanted to contribute something and found the open "Rewrite tuxvdmtool to userspace i2c" Task on your Help Wanted page. I rewrote the needed cd321x driver logic in rust using the i2cdev crate as proposed.
18:49Bjarne[m]In the [issue](https://github.com/AsahiLinux/tuxvdmtool/issues/1) there is also the point of building a single tools with multiple backends. How should this look like?
18:50svenplease try to limit edits on the matrix side, this channel is mainly on irc and edits are very disturbing here
18:51fl0_id___yuka, mmmh, I also thought the micro-usb of cs should me able to connect to m1n1, like to use before usb is up. as the readme says "It's main use it to allow interacting with the Asahi m1n1 firmware, such as tethered booting"
18:53ShizBjarne[m]: probably some kind of standard 'backend API' (rust trait) along with a way to enumerate them
18:53fl0_id___and do let me know if you figure out to force-print sth in early kernel boot... I rather suspect it doesn't work because of some memory access weirdness like trying to acccess high memory regions
18:53Shizi2cdev could be one backend, the IOKit HPM interface for macOS could be another
18:53hdbngr has joined #asahi-dev
18:57Bjarne[m]Shiz: Thanks, i will take a look at it! Should i propose a userspace version for tuxvdm meanwhile?
18:57Shizdefinitely, can't hurt to take it one at a time :)
18:58svenyeah, that would already be useful :)
19:00jannauyes, please do. that was the main objective. I just realized that macvdmtool and the i2c userspace tuxvdmtool would do mostly the same
19:07hdbngr has quit [Ping timeout: 480 seconds]
19:13pb17 has quit [Ping timeout: 480 seconds]
19:21pb17 has joined #asahi-dev
19:31hdbngr has joined #asahi-dev
19:40hdbngr has quit [Ping timeout: 480 seconds]
19:49jannauinstall seems to work now after upgrading the main os to 15.7. I suspect the installer wanted newer system firmware/recovery and was not able install it itself
19:56jannauyes, installer went through
19:56sventbt is still broken, found out i missed to set the orientation but that didn't fix anything :(
20:21jannauit broke while updating from a linux 6.13 based branch to 6.18 / 6.19-rc / linux-next? are tipd and atc-phy are identical? any relevant changes in dart?
20:22bigsilicon has quit [Quit: bigsilicon]
20:25svenlast time it definitely worked was 6.10
20:25sveni *think* it also worked with 6.13
20:25svengonna compare atpphy traces tomorrow and see if there's a difference
20:26sven*atcphy
20:32hdbngr has joined #asahi-dev
20:35__nick__ has quit []
20:37___nick___ has joined #asahi-dev
20:38___nick___ has quit []
20:40___nick___ has joined #asahi-dev
20:43fl0_id___ has quit [Ping timeout: 480 seconds]
20:45Bjarne[m]and i think my message might have been to long for matrix to irc?
20:45Bjarne[m]Sorry, I haven't used IRC a lot...
20:46ShizI can test it tomorrow
20:47hdbngr has quit [Ping timeout: 480 seconds]
20:48svenlong messages are only slightly inconvenient but usually fine. only editing is what ends up very awkward on the irc side
20:49hdbngr has joined #asahi-dev
20:50Bjarne[m]sven: Yes, sorry for that one again ^^
20:50svenno worries :)
20:50Bjarne[m]Shiz: Cool, thanks!
20:57hdbngr has quit [Ping timeout: 480 seconds]
20:57pb17 has quit [Ping timeout: 480 seconds]
21:01Yushi_ has joined #asahi-dev
21:04___nick___ has quit [Remote host closed the connection]
21:07pb17 has joined #asahi-dev
21:08Yushi has quit [Ping timeout: 480 seconds]
21:09Yushi_ is now known as Yushi
21:29hdbngr has joined #asahi-dev
22:52pb17 has quit [Ping timeout: 480 seconds]
23:01ethantwardy has quit [Quit: WeeChat 4.4.2]
23:02pb17 has joined #asahi-dev
23:07ethantwardy has joined #asahi-dev
23:09ethantwardy has quit []
23:14ethantwardy has joined #asahi-dev
23:15ethantwardy has quit []
23:18__Tetris__1 has joined #asahi-dev
23:18__Tetris__ has quit [Ping timeout: 480 seconds]
23:18ethantwardy has joined #asahi-dev
23:18ethantwardy has quit []
23:24ethantwardy has joined #asahi-dev
23:25ninelore has quit [Ping timeout: 480 seconds]
23:26__Tetris__1 has quit [Ping timeout: 480 seconds]
23:27maor26 has quit [Ping timeout: 480 seconds]
23:28hdbngr has quit [Ping timeout: 480 seconds]
23:31ethantwardy has quit [Quit: WeeChat 4.4.2]
23:34ethantwardy has joined #asahi-dev
23:36hdbngr has joined #asahi-dev
23:44ethantwardy has quit [Quit: WeeChat 4.4.2]
23:50ethantwardy has joined #asahi-dev
23:55ninelore has joined #asahi-dev
23:55__Tetris__ has joined #asahi-dev
23:58hdbngr has quit [Ping timeout: 480 seconds]
…