#
…

2026-10-06

00:18Mayanico13 has left #asahi-dev [Leaving]
00:25meow676767676 has joined #asahi-dev
00:31bgb has quit [Quit: WeeChat 4.10.1]
00:32bgb has joined #asahi-dev
00:33bgb has quit []
00:35bgb has joined #asahi-dev
00:36lion328 has joined #asahi-dev
01:24linkmauve has quit [Remote host closed the connection]
01:28linkmauve has joined #asahi-dev
02:00yurix has quit [Remote host closed the connection]
02:00yurix has joined #asahi-dev
02:21tobhe_ has joined #asahi-dev
02:26betoissues has quit [Quit: bye bye.]
02:26bgb has quit [Remote host closed the connection]
02:27betoissues has joined #asahi-dev
02:28tobhe has quit [Ping timeout: 480 seconds]
02:29bgb has joined #asahi-dev
03:14mripard has quit [Remote host closed the connection]
03:15mripard has joined #asahi-dev
03:18voidah has quit [Ping timeout: 480 seconds]
03:29betoissues has quit [Quit: bye bye.]
03:30voidah has joined #asahi-dev
03:34betoissues has joined #asahi-dev
03:41nora_ has joined #asahi-dev
03:43nora has quit [Ping timeout: 480 seconds]
04:07MichaelReeves[m]myria: looked through ur PMU pre-review branch, it looks really good I just left a couple comments :)
04:07myriaI saw! thank you, sorry was in a lecture
04:08MichaelReeves[m]oh that's okay dont worry
04:09MichaelReeves[m]just wanted to let you know in case you didnt
04:16voidah has quit [Ping timeout: 480 seconds]
04:24voidah has joined #asahi-dev
04:29eluks has quit [Remote host closed the connection]
04:30eluks has joined #asahi-dev
04:32voidah has quit [Ping timeout: 480 seconds]
04:40voidah has joined #asahi-dev
04:49voidah has quit [Ping timeout: 480 seconds]
04:57voidah has joined #asahi-dev
05:05voidah has quit [Ping timeout: 480 seconds]
05:11voidah has joined #asahi-dev
05:16myriaJust fixed up all your feedback, thank you for it! That'll teach me to rebase on no sleep :P
05:19myriaDo you have the M3 series errata fix ready? I can look over it before you send it as well, if you want.
05:20voidah has quit [Ping timeout: 480 seconds]
05:27hello0092 has joined #asahi-dev
05:27hello0092 has quit []
05:27Munzzyy[m] has joined #asahi-dev
05:27hello0092 has joined #asahi-dev
05:28voidah has joined #asahi-dev
05:37voidah has quit [Ping timeout: 480 seconds]
05:41MichaelReeves[m]I can put it up on GitHub if you'd like, but I didn't initially because its small
05:42null_ is now known as null
05:43bennofs3 has quit [Ping timeout: 480 seconds]
05:44MichaelReeves[m]i split it into two commits also so one of them (the MIDR) is From: you
05:45voidah has joined #asahi-dev
05:47bennofs3 has joined #asahi-dev
05:48gargantua has joined #asahi-dev
05:54voidah has quit [Ping timeout: 480 seconds]
05:54myriaDo you want to add the other MIDR will you're at it, so that we don't have to in future?
05:55myriaOtherwise, looks good from the quick glance I had at it
05:57MichaelReeves[m]thanks! I don't think the maintainers will like unused/dead code so they probably have to wait for PMU where they actually have a user
05:57MichaelReeves[m]but I might be assuming wrong?
05:59myriaEhhhh, true. I thought it might be okay because it was just MIDR definitions and they're not going to change, but yeah if we don't even support them then probably no point
05:59myriaOh well, I'll keep them around for when we need the rest
06:00sofus_ has joined #asahi-dev
06:00mattia013 has quit [Remote host closed the connection]
06:01mattia013 has joined #asahi-dev
06:02voidah has joined #asahi-dev
06:02sofus has quit [Ping timeout: 480 seconds]
06:14voidah has quit [Ping timeout: 480 seconds]
06:23voidah has joined #asahi-dev
06:32voidah has quit [Ping timeout: 480 seconds]
06:39voidah has joined #asahi-dev
06:43myriaFor the M3-M6 PMU events series I have, do you reckon I should wait till v7.4-rc1 comes out, then send it including M4/A18 bindings
06:43myriaOr send it soon-ish and add them later/depend on the changeid of the M4/A18 patches?
06:46chadmed7.4-rc1 is only in a couple of weeks so maybe just wait
06:49voidah has quit [Ping timeout: 480 seconds]
06:52hardy94 has joined #asahi-dev
06:58voidah has joined #asahi-dev
07:06voidah has quit [Ping timeout: 480 seconds]
07:12sofus_ is now known as sofus
07:16voidah has joined #asahi-dev
07:19MichaelReeves[m]I like the regmap and maple tree maintainers they reply so fast!
07:22myriaIndeed!
07:25voidah has quit [Ping timeout: 480 seconds]
07:33voidah has joined #asahi-dev
07:41voidah has quit [Ping timeout: 480 seconds]
07:46MichaelReeves[m]I'm trying to test just rc6 + the patch before I send it but getting qemu + KVM to work in my mini rootfs is annoying
07:46MichaelReeves[m](send errata KVM patch)
07:47j`eyMichaelReeves[m]: you could look into kvmtool, it's a lot simpler than qemu
07:47MichaelReeves[m]oooh thanks ill have to check it out
07:50voidah has joined #asahi-dev
07:54LarstiQnoopwafel: welcome back! Oh that sounds closer than I expected. For rendering in m1n1 it made me think of running intro's like https://github.com/MindlapseDemos/256boss
07:58voidah has quit [Ping timeout: 480 seconds]
08:06voidah has joined #asahi-dev
08:11gargantua has quit [Ping timeout: 480 seconds]
08:12hello0092 has quit [Remote host closed the connection]
08:12jdragoyuka: (and everyone else) your help with Hel3zi so far has also been helping me in the background, like a LOT, so genuinely thank you so much
08:13hello0092 has joined #asahi-dev
08:13jdragoand also your help with allusivedev to an extent!
08:15voidah has quit [Ping timeout: 480 seconds]
08:15jdragoyou got me further in a few days than i managed to get myself over a few weeks, so seriously thank all of yall so much
08:23voidah has joined #asahi-dev
08:24hello0092 has quit [Ping timeout: 480 seconds]
08:29bennofs3 has quit [Ping timeout: 480 seconds]
08:31voidah has quit [Ping timeout: 480 seconds]
08:39voidah has joined #asahi-dev
08:46MichaelReeves[m]oh just realised need to extent KVM patchset to 3 patches, M3 needs to be added to vgic_v3_broken_seis as well
08:46MichaelReeves[m]s/extent/extend
08:46MichaelReeves[m]after that all working fine on upstream
08:46MichaelReeves[m]will send soon
08:47myriawoohoo!
08:48voidah has quit [Ping timeout: 480 seconds]
08:50chadmed has quit [Read error: Connection reset by peer]
08:50chadmed has joined #asahi-dev
08:51chadmedsofus: https://paste.debian.net/hidden/56cb917f i think we have a memory leak
08:55voidah has joined #asahi-dev
08:57bennofs3 has joined #asahi-dev
08:58jannau_"Your requested paste entry '56cb917f' could not be found "
08:59sofushuh, i could open it
08:59chadmedyeah opens fine for me
08:59chadmedmaybe its replicating :P
09:01yukajdrago if you found anything missing in the docs during this, you could open a PR to improve them
09:02jannau_works now
09:03MichaelReeves[m]minor nit but do I add "Hi," to cover letter start? I normally don't but I've seen others do it and maybe it seems nicer?
09:03chadmedi always "hey folks"
09:03chadmedit is a letter after all
09:04chadmedsofus: have you seen the __free cleanup annotation?
09:04chadmedyou should use those instead of explicit kfrees and gotos where possible to make sure that allocated memory is freed as soon as it is descoped
09:06jannau_the paste doesn't look like memory leak but a too large allocation for kmalloc(), if the allocation size is correct, it should use vmalloc()
09:08MichaelReeves[m]iirc someone submitted an AI PR for that a while back?
09:08MichaelReeves[m]idk if that's different though
09:09chadmedmy eyes just glaze over when i see this bullshit
09:09chadmedcommit message longer than the patch
09:11chaos_princessnuke the stupid comment and merge it
09:12chaos_princesslike, didn't the whole "literate programming" fad die a while ago, why are we bringing it back
09:12integralpilot has joined #asahi-dev
09:12voidah has quit [Ping timeout: 480 seconds]
09:14chadmedthe problem is the terabytes and terabytes of babbys first hellorld program tutorials that the robot trains on
09:15chadmedand of course uni course material being waaaay too verbosely commented because the whole point is to teach noobs what things do
09:15chadmedso of course when noobs get a hold of the robot and go "make apple go brrr make no mistakes" they dont know theyre supposed to get rid of the verbose slop comments /* Assign value to variable */
09:16sofusits done totally differently in my ml version. I can update the whole thing to use kvzalloc instead but thats likely not necessary if 615 gets merged
09:16chadmedim not in favour of merging 631
09:16chadmedid rather merge the actual author's fix :P
09:17chadmedor just wait until we rebase on whatever kernel version ends up with avd in it upstream
09:17chaos_princessnp then, more commenting on the fact that 631 is technically a valid fix, just wrapped in layers of... whatever that is
09:18sofusnot really tho, it would also need to do the same in avd-drv
09:18sofusoh wait sorry
09:19mattia013 has quit [Quit: mattia013]
09:19mattia013 has joined #asahi-dev
09:20voidah has joined #asahi-dev
09:26noopwafelLarstiQ: yes it's really not that huge a diff in the end on kernel side :) current diff (and I did end up fixing it btw, \o/) is <5k lines and most of that is basically churn like added ::ver. the refactor of much of initdata to m1n1 helps a lot.
09:26noopwafeloh I already said the latter :P but I cannot praise enough
09:28voidah has quit [Ping timeout: 480 seconds]
09:28chaos_princessmaybe the vendors were onto something with "just put code into the firmware instead of having to upstream it" :P
09:36voidah has joined #asahi-dev
09:44voidah has quit [Ping timeout: 480 seconds]
09:53voidah has joined #asahi-dev
09:58MichaelReeves[m]ok so turns out the "new errata" I made in my patch was already part of a larger "apple CPUs broken SEIS" thing that all M1 and M2 are in
09:58MichaelReeves[m]it just needs to be adding M3 to that
09:59MichaelReeves[m]and while we're doing it might as well add m4, m5, m6 and the other MIDRs myria has ready
09:59MichaelReeves[m]thankfully I was doing some last checks before I send
10:00MichaelReeves[m]I didn't notice because that other thing was making the changes in a completely different place than I expected (and added my patch) and also its naming didnt make it sound like it included that thing as well
10:00MichaelReeves[m]anyway... at least I know now
10:01voidah has quit [Ping timeout: 480 seconds]
10:02jannau_MichaelReeves[m]: do not add SoCs nobody has tested
10:03MichaelReeves[m]ok i was just saying because myria mentioned it might be worth doing that before but probably good idea not to then
10:03MichaelReeves[m]so just m3 or has anyone tested m4 PMU/KVM (its all worked into the one generic apple cpus errata)
10:03jannau_ is now known as jannau
10:05MichaelReeves[m]from Myria's branch I don't think someone's done m4 yet so I'll just keep m3 for now I think
10:06yukasomeone has tested kvm on M4, but I'm afraid they're not in here
10:06yukaI expect it will be similar to M3
10:06yukaI should test it to confirm
10:07voidah has joined #asahi-dev
10:13MichaelReeves[m]would you like to test? (if not cause ur busy with other stuff it's fine because its harmless to just send m4 later, it's tiny patch). if so, thanks & I'll wait before sending series.
10:15voidah has quit [Ping timeout: 480 seconds]
10:17myriaMichaelReeves[m]: I missed it too, dw. Probably related to some of the funky PMU stuff they do, and how we emulate PMUv3 on it for guests
10:23voidah has joined #asahi-dev
10:36voidah has quit [Ping timeout: 480 seconds]
10:37anon_ has joined #asahi-dev
10:38anon_Hello, I bet you guys have read my misplace post in #asahi about m5 nvme.
10:38anon_Any comments?
10:39anon_ has quit [Remote host closed the connection]
10:42MichaelReeves[m]did keyboard just work with adding it to DT? there were no dchid changes?
10:43chadmedthey left again, hopefully they read the logs
10:44voidah has joined #asahi-dev
10:48integralpilot has quit [Ping timeout: 480 seconds]
10:52voidah has quit [Ping timeout: 480 seconds]
10:53yukaMichaelReeves[m] would gladly test
10:53yukawhere can I find the patch?
10:54MichaelReeves[m]thank you! can you try KVM without it and then what should happen is either EL2 SError or it hangs. and then apply the patch (will send soon) and it should work fine
10:56myriaWant me to test the patch on M3 max too?
10:56MichaelReeves[m]yes please thanks
11:01voidah has joined #asahi-dev
11:03yukaM3 Max is very easy for me to test, as it's my main device and already has everything installed. On M4, I'll need a few minutes to put together a setup where I can boot a more complete system with qemu and networking.
11:03yukaI know M3 Max is weirdly broken. The guest does start to some degree but then hangs (nixos tests never reach their backdoor target).
11:04yukawith nested virt, the host also sometimes hangs
11:05MichaelReeves[m]yep, that's the problem this should fix, hopefully
11:06Mayanico13 has joined #asahi-dev
11:09bennofs3 has quit [Ping timeout: 480 seconds]
11:13voidah has quit [Ping timeout: 480 seconds]
11:17bennofs3 has joined #asahi-dev
11:23voidah has joined #asahi-dev
11:23Takeshi has joined #asahi-dev
11:24meow676767676 has quit [Quit: Ping timeout (120 seconds)]
11:24meow676767676 has joined #asahi-dev
11:25integralpilot has joined #asahi-dev
11:25integralpilotabout to run first test of asahi-6.18 + those android patches + m3 backports, let's see how it goes!
11:26integralpilotif 4K works that would be awesome but i feel like there's gonna be some bug
11:30chadmedholy shit
11:31yukaamazing
11:31voidah has quit [Ping timeout: 480 seconds]
11:32chadmeddoes the patchset include the elf loader magic they were talking about
11:32chadmedor its just not required
11:32integralpilotit shows 4096 in sysconf and AT_PAGESZ and i made it do a bunch of nonsense like messy forks and WORKS etc. and it all works
11:32integralpilot(was too excited about it working lol and got words mixed up)
11:33gifti has joined #asahi-dev
11:34integralpilotchadmed: yes it does
11:34chadmedyeah ok cool i figured it must
11:34chadmedwe are so back
11:34j`eywhat's "those android patches"
11:34integralpilotrunning 4K page size programs natively
11:34chadmedj`ey: they actually did the thing
11:35j`eyjannau: danke
11:35integralpilotone more thing to test though, the ELF i used had 64K aligned segments (but i ran it in 4K mode to test), should probably try layout that needs 4K alignment next
11:35integralpilotbut pretty sure it will work
11:35chadmedtldr elf aligned to 4k, loader will mark it as such and the granularity will be changed to 4k for that process only, and then switched back on context switches and such
11:35chadmedits pretty cool
11:37jannauintegralpilot: can you push your tree so someone can test on M1/M2 with FEX and GPU drivers
11:37anon_ has joined #asahi-dev
11:38j`eyI don't see it, but is this running all userspace processes with 4k? hard to scan without a full diff
11:38chadmedno, its per process
11:38chadmedthe holy grail doctor jones
11:39integralpilotjannau: yep will do soon
11:39jannauwhen that works it would make sense to contact the authors reports that it works and enquire about upstream plans
11:39chadmedhopefully the rebase isnt too painful, i want to do some benchmarks (assuming it works)
11:39voidah has joined #asahi-dev
11:40integralpilotuh oh the one that strictly needs 4K has some errors... its ok though i just think its a simple fix
11:41jannaurebasing and upstreaming will be painful. upstreaming will be hard because it covers only one specific case (4k processes on a 16k arm64 kernel)
11:41chadmedi dont think it will ever fly upstream unless it is made generic across pagesizes
11:41chadmedat least for all system size > process size cases
11:42chadmedthe other direction doesnt matter aiui, a 64k aligned thing will work fine on a 16k system
11:47Takeshi has quit [reticulum.oftc.net helix.oftc.net]
11:47sofus has quit [reticulum.oftc.net helix.oftc.net]
11:47eluks has quit [reticulum.oftc.net helix.oftc.net]
11:47mripard has quit [reticulum.oftc.net helix.oftc.net]
11:47tobhe_ has quit [reticulum.oftc.net helix.oftc.net]
11:47yurix has quit [reticulum.oftc.net helix.oftc.net]
11:47linkmauve has quit [reticulum.oftc.net helix.oftc.net]
11:47covkie[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47yuexiayouren[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47forsakendurian[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47httperry[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47Umut[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47SnipeX_[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47funta[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47loikg[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47jonaias[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47kereru[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47IsaacSmith[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47Tachyonray[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47MalpasoMan[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47chanelno19[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47faranith[mds] has quit [reticulum.oftc.net helix.oftc.net]
11:47LuisHenriques[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47elisaado has quit [reticulum.oftc.net helix.oftc.net]
11:47sofus[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47tumblingweed has quit [reticulum.oftc.net helix.oftc.net]
11:47sea[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47xeon[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47ConnorAlecks[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47Eighth_Doctor has quit [reticulum.oftc.net helix.oftc.net]
11:47rhlvista[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47BigDaddy[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47neobrain[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47jules[m]1 has quit [reticulum.oftc.net helix.oftc.net]
11:47mungasi2000[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47kedde[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47hex86[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47eluks has joined #asahi-dev
11:47yurix has joined #asahi-dev
11:47jobe[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47user1[m] has quit [reticulum.oftc.net helix.oftc.net]
11:47Nick[m]1 has quit [reticulum.oftc.net helix.oftc.net]
11:47jvoisin has quit [reticulum.oftc.net helix.oftc.net]
11:47noopwafel has quit [reticulum.oftc.net helix.oftc.net]
11:47mjolnir[mds] has quit [reticulum.oftc.net helix.oftc.net]
11:47averne_ has quit [reticulum.oftc.net helix.oftc.net]
11:47mixi has quit [reticulum.oftc.net helix.oftc.net]
11:47tobhe has joined #asahi-dev
11:47averne has joined #asahi-dev
11:47mripard has joined #asahi-dev
11:48voidah has quit [Ping timeout: 480 seconds]
11:48anon_For the M5 nvme init problem, I'm not 100% sure resetting is not viable at all because it's another person(ofc a real guy) who did all the experiments. But still, you can init nvme in m1n1, and re-adopt them in nvme-apple.
11:48jvoisin has joined #asahi-dev
11:48mixi has joined #asahi-dev
11:49jvoisin has quit []
11:49sofus has joined #asahi-dev
11:49jvoisin has joined #asahi-dev
11:49sofus[m] has joined #asahi-dev
11:50sofus[m] has quit [reticulum.oftc.net helix.oftc.net]
11:50sofus has quit [reticulum.oftc.net helix.oftc.net]
11:50mixi has quit [reticulum.oftc.net helix.oftc.net]
11:50mixi has joined #asahi-dev
11:50integralpilothuh whats that netsplit thing, havent seen it before
11:50sofus has joined #asahi-dev
11:51integralpilotbtw narrowed it down its smth with dynamic linking. static 4K passes. im 99% sure i see the issue so it should be fixed next boot
11:52linkmauve has joined #asahi-dev
11:53chadmeddont snipe yourself into fixing too many things
11:53chadmedlet's find out what google's plans are for this first
11:53chadmedif they cbf doing the upstreaming work themselves maybe then we take a second look at it
11:55user1[m] has joined #asahi-dev
11:56integralpilotthanks, i wont
11:56integralpilotif this fix im trying now doesnt work ill probably just upload the branch and go to bed
11:56integralpilotlikewise, if it does work, same
11:57LuisHenriques[m] has joined #asahi-dev
11:57anon_ has quit [Remote host closed the connection]
11:57noopwafel has joined #asahi-dev
11:57voidah has joined #asahi-dev
11:58loikg[m] has joined #asahi-dev
11:58jonaias[m] has joined #asahi-dev
11:58yukatesting the KVM patch on M3 Max...
11:59kereru[m] has joined #asahi-dev
12:00Eighth_Doctor has joined #asahi-dev
12:01IsaacSmith[m] has joined #asahi-dev
12:03rhlvista[m] has joined #asahi-dev
12:03elisaado has joined #asahi-dev
12:04Tachyonray[m] has joined #asahi-dev
12:05yukadoes not seem to work (yet)
12:06tumblingweed has joined #asahi-dev
12:06yukaguest: [ 24.003964] watchdog: CPU0: Watchdog detected hard LOCKUP on cpu 1
12:06integralpilotcan you try the other one
12:06integralpilotthat i scrapped but was about to send
12:06integralpilothere ill get the link again
12:06integralpilotmaybe i wasnt getting mixed up after all and that's the right one
12:07SnipeX_[m] has joined #asahi-dev
12:07integralpilotmeanwhile in other news after my tiny fix (+11/-5) the dynamic 4K program is WORKING
12:07integralpilotits doing all the nasty stuff (fork, COW, etc.) i tested before so that's good
12:07integralpiloti will upload branch now
12:08yukabuilding...
12:08integralpilotname needs to be something unappealing so people dont ship it early (ofc, that wont work but anyway)
12:09bennofs3 has quit [Ping timeout: 480 seconds]
12:09Hel3zi has joined #asahi-dev
12:09Hel3ziyuka: hey! Getting the internal keyboard to work on M4 Pro takes quite a bit of sweat, doesn't it? I got dockchannel-hid initialized on t6040, but STM/AFE looks like it's held in reset via SMC. Did you manage to bring up topcase on M4 yet?
12:09Hel3zi has quit [Remote host closed the connection]
12:09ConnorAlecks[m] has joined #asahi-dev
12:09yukaHel3zi I don't have any MacBook with M4
12:09yukaonly an Mac mini
12:10yukaBut there have been successful attempts
12:10yukaI think this one was working, but it's using the downstream dockchannel-hid
12:10yukaat least was definitely working on the M5 Pro MacBook Pro
12:11yukaand here someone managed to do it for the base M4: https://github.com/AsahiLinux/linux/pull/567
12:11Hel3zi has joined #asahi-dev
12:11yuka(also using downstream dockchannel-hid)
12:12integralpilotoh no does upstream have anything not right?
12:12integralpiloti tried my best but happy to fix if anything needed
12:12yukaNo, at that point there was simply no draft
12:12faranith[mds] has joined #asahi-dev
12:13yukaHel3zi: When testing, you can always let m1n1 write the smc gpio pin
12:13covkie[m] has joined #asahi-dev
12:13yukasee src/dcp.c:
12:13yuka"smc_write_u32(smc, dcp->dp2hdmi_pwr_gpio, 0x800001);"
12:13yukasimilar to that
12:13yukaadd it to kboot
12:14yukathey should work the same on M4 Pro / M5 / M5 Pro
12:14bennofs3 has joined #asahi-dev
12:14voidah has quit [Ping timeout: 480 seconds]
12:15yukaintegralpilot: writing the M4+ device trees for upstream dockchannel-hid is on my todo list
12:15integralpilotif you have the downstream ones, i can do it for you if you'd like.
12:16yukaSee the links above
12:17integralpilotcool, will try tomorrow
12:17BigDaddy[m] has joined #asahi-dev
12:17mjolnir[mds] has joined #asahi-dev
12:17integralpilotjust need to upload branch and then bed, tryna make sure its all right
12:18forsakendurian[m] has joined #asahi-dev
12:20yukaintegralpilot: is it possible that the midr needs to be added to ‎has_vgic_v3 in arch/arm64/kvm/vgic/vgic-v3.c as well?
12:20integralpilotooh maybe thanks. sorry its getting too late and i might be stuffing up all these patches
12:21integralpilotcan you please try it?
12:21yukafirst trying the other patch you sent since I've now already built that
12:21integralpilotthank you for helping with all this!
12:22Nick[m]1 has joined #asahi-dev
12:22yukathank you for doing it!
12:23yukabefore you send it upstream let's also add and test the A18 Pro Tahiti MIDR part numbers
12:24voidah has joined #asahi-dev
12:24jules[m]1 has joined #asahi-dev
12:27mungasi2000[m] has joined #asahi-dev
12:28kedde[m] has joined #asahi-dev
12:29hex86[m] has joined #asahi-dev
12:30yukawith the second patch it fails earlier
12:33voidah has quit [Ping timeout: 480 seconds]
12:35Umut[m] has joined #asahi-dev
12:36yukatesting the first patch + ‎has_vgic_v3 now
12:37Hel3zi has quit [Remote host closed the connection]
12:42xeon[m] has joined #asahi-dev
12:42voidah has joined #asahi-dev
12:46yuexiayouren[m] has joined #asahi-dev
12:47neobrain[m] has joined #asahi-dev
12:48sea[m] has joined #asahi-dev
12:51voidah has quit [Ping timeout: 480 seconds]
12:51integralpiloti chose warthog cause i was thinking what animals are associated with messiness
12:51integralpilotanyway testing on M1/M2 or reproduction on M3 would be appreciated :)
12:52integralpilot(the way i set it up was asahi-6.18 -> m3-backports -> warthog, so it can work with m3)
12:52tryhard[m]hippo is more messy
12:52httperry[m] has joined #asahi-dev
12:53chanelno19[m] has joined #asahi-dev
12:53MalpasoMan[m] has joined #asahi-dev
12:53funta[m] has joined #asahi-dev
12:55integralpiloti wonder if games work out of the box, i feel like the answers gonna be no but only one way to find out (and i cant rly on m3)
12:55integralpilotmaybe v2 once its better can be hippo :)
12:56jobe[m] has joined #asahi-dev
12:57chadmedi will try to dump this on top of 7.1 on the weekend or something and see whats up
12:57chadmedits probably gonna be painful but whatever
12:57sofus[m] has joined #asahi-dev
13:00integralpilotill try to rebase tmrw and see if it works
13:00integralpilotanyway goodnight!
13:00integralpilot has quit [Quit: Lost terminal]
13:01voidah has joined #asahi-dev
13:53yukathis works for me on M3 Max: https://termbin.com/44r8m
13:53yukanow testing nested virt
13:54myriaHmmm, only concern I have with that is what having `impdef_pmuv3` implies if M3/M4 doesn't have the PMU driver
13:55myriaBut good to hear that it fixes it!
14:04myriaThat's the rest of the MIDRs I had
14:04yukaI just left home and didn't take the MB Neo
14:04yukabut I did take the M5 Pro
14:06yukatrying to get the libvirtd nixos test to build with nested virt
14:06yukait requires some special qemu flags
14:06myriaThat sounds like fun
14:07myriaI'm not sure, but only M3+ has nested virt on Apple, so not sure if it's the same here
14:08myriaThat, and we do funky emulation of the PMU stuff that I'm not sure will work in a nested virt setup... though I guess if it's a regular cpu in the kvm it won't notice...
14:09yukaThe PMU emulation stuff requires adding the new PMU events and more bits per event thing?
14:09yukafor M3
14:09yukaOr is the emulation not built on top of the Apple PMU thing
14:10myriaIt is built on that. From my quick reading, it seems that apple allows writes to some registers to trap, which we catch and then set the events in the host kernel
14:10yuka> watchdog: BUG: soft lockup - CPU#1 stuck for 24s! [txg_sync:1092]
14:11j`eymyria: M2 have nested virt
14:11yukaI need to make it use a 16K pagesize kernel for the nixos test because 4K pagesize guests are not supported with nested virt
14:11yuka> kvm [1]: PAGE_SIZE not supported at Stage-2, giving up
14:11yukabut also with this it locks up....
14:11myriaBut it should (hopefully) just require us setting up the event info for apple PMU, and then it should all work?
14:12myriaj`ey: in macOS it doesn't. Win for linux there!
14:14myriayuka: annoying. I seem to have `nixosTest.libvirtd` running fine right now on my M2, will let you know if it completes
14:14myriaHaven't run into any issues with pages/softlocks
14:15yukaIt wil just run the inner layer with tcg
14:15voidah has quit [Ping timeout: 480 seconds]
14:16yukaUnless you do sth like this (will send diff in 5min)
14:16yukaSoftware emulation
14:16yukaCheck the boot logs for kvm
14:16yukaIt will say no hyp mode supported
14:21myriaYep, found it
14:21yukaqyliss has def run spectrum on M2 with nested virt before, but with some mainline + nv patches frankenstein kernel
14:22yukaOk, so
14:22yukaI added "kvm-arm.mode=nested" on the host
14:23yukaalso it locked up my host shortly after
14:23yukaso nested virt is still heavily broken
14:23voidah has joined #asahi-dev
14:24myriaUnsurprising
14:24myriaI'm going to sleep now, best of luck with your testing!
14:24yukagood night!
14:29Hel3zi has joined #asahi-dev
14:29Hel3zi has quit [autokilled: This host violated network policy. Mail support@oftc.net if you think this is in error. (2026-10-06 14:29:22)]
14:29Hel3zi has joined #asahi-dev
14:29Hel3ziUnfortunately, I haven't managed to get the keyboard working yet :(
14:32voidah has quit [Ping timeout: 480 seconds]
14:34Hel3zi has quit [Remote host closed the connection]
14:37yukahel3zi I will probably be able to write this up and test it on thursday
14:38yukafor M4 Pro
14:40voidah has joined #asahi-dev
14:41hardy94 has quit [Quit: hardy94]
14:41hardy94 has joined #asahi-dev
14:41bennofs3 has quit [Ping timeout: 480 seconds]
14:44bennofs3 has joined #asahi-dev
14:49hardy94 has quit [Ping timeout: 480 seconds]
14:49Takeshi has joined #asahi-dev
15:02Hel3zi has joined #asahi-dev
15:02Hel3ziThat'd be great :) I'll need to test it afterwards, right?
15:03yukaI will have a machine available to test it on thursday
15:03yukabut you can of course still test it
15:04yukaanon_: you already left but in case you read the logs: m1n1 can successfully init and shutdown the nvme on t6050, so it's some linux thing I guess
15:04yuka(and also re-init after shutdown)
15:08Hel3ziyuka: Oh nice, will it be an M4 Pro MacBook? For how long will you have it?
15:08yukaone evening
15:10Hel3ziThen we'll have to get a lot done :) Do you know roughly around what time (and timezone) that'll be?
15:11yukaguessing a bit but ~18:00-~23:00 central european
15:12Hel3ziGot it! I'll be around and ready to test then. Thanks and see you on Thursday!
15:20Hel3zi has quit [Remote host closed the connection]
15:25voidah has quit [Ping timeout: 480 seconds]
15:27Hel3zi has joined #asahi-dev
15:27Hel3ziWow, they're already running it on the M5...
15:29yukahel3zi: one of the Asahi people here already got an M6, so ;)
15:32voidah has joined #asahi-dev
15:41voidah has quit [Ping timeout: 480 seconds]
15:43Hel3zireally?! It feels like a scam
15:49voidah has joined #asahi-dev
15:53okb has joined #asahi-dev
15:57voidah has quit [Ping timeout: 480 seconds]
16:05voidah has joined #asahi-dev
16:13Hel3zi has quit [Remote host closed the connection]
16:14chadmed has quit [Read error: No route to host]
16:14chadmed has joined #asahi-dev
16:16chadmed has quit []
16:16chadmed has joined #asahi-dev
16:16chadmed has quit [Read error: Connection reset by peer]
16:17chadmed has joined #asahi-dev
16:19hello0092 has joined #asahi-dev
16:21hello0092 has quit []
16:21hello0092 has joined #asahi-dev
17:27okb has quit [Ping timeout: 480 seconds]
17:36okb has joined #asahi-dev
17:37hello0092 has quit [Quit: Konversation terminated!]
17:37hello0092 has joined #asahi-dev
18:08hub2 has quit [Quit: Ping timeout (120 seconds)]
18:09hub2 has joined #asahi-dev
18:21foltz58 has joined #asahi-dev
18:24hardy94 has joined #asahi-dev
18:25foltz5 has quit [Ping timeout: 480 seconds]
18:25foltz58 is now known as foltz5
18:44okb has quit [Ping timeout: 480 seconds]
18:51hello0092 has quit [Quit: Konversation terminated!]
18:51hello0092 has joined #asahi-dev
18:53runxiyu_ has quit []
19:14okb has joined #asahi-dev
19:27hardy94 has quit [Quit: hardy94]
19:33foltz54 has joined #asahi-dev
19:38foltz5 has quit [Ping timeout: 480 seconds]
19:38foltz54 is now known as foltz5
19:51lena6 has joined #asahi-dev
19:51lena6 has quit [Remote host closed the connection]
19:52lena6 has joined #asahi-dev
19:57okb has quit [Remote host closed the connection]
19:57okb has joined #asahi-dev
20:01jcbm_ has joined #asahi-dev
20:04jcbm has quit [Remote host closed the connection]
20:05voidah has quit [Remote host closed the connection]
20:05jcbm_ is now known as jcbm
20:05voidah has joined #asahi-dev
20:15lena6 has quit [Read error: Connection reset by peer]
20:15lena6 has joined #asahi-dev
20:54Mayanico13 has quit [Quit: Leaving]
20:54voidah has quit [Ping timeout: 480 seconds]
21:01voidah has joined #asahi-dev
21:05okb has quit [Ping timeout: 480 seconds]
21:25voidah has quit [Ping timeout: 480 seconds]
21:48voidah has joined #asahi-dev
21:56voidah has quit [Ping timeout: 480 seconds]
22:02hardy94 has joined #asahi-dev
22:19bgb has quit [Remote host closed the connection]
22:19bgb has joined #asahi-dev
22:22hardy94 has quit [Remote host closed the connection]
22:35elara has quit [Ping timeout: 480 seconds]
23:01elara has joined #asahi-dev
23:26elara has quit [Ping timeout: 480 seconds]
23:31mattia013_ has joined #asahi-dev
23:36mattia013 has quit [Ping timeout: 480 seconds]
23:36mattia013_ is now known as mattia013
23:36hello0092 has quit [Quit: Konversation terminated!]
23:44rany has quit [Remote host closed the connection]
23:44rany has joined #asahi-dev
23:45elara has joined #asahi-dev
…