2026-11-09
00:05Lyude has joined #asahi-dev
01:05sinan[m] has joined #asahi-dev
01:08EnnuiLangeweile[m] is now known as EnnuiL
01:08chadmed has quit [Remote host closed the connection]
01:08chadmed has joined #asahi-dev
01:10chadmed has quit [Read error: Connection reset by peer]
01:10chadmed has joined #asahi-dev
01:10chadmed has quit [Read error: Connection reset by peer]
01:11chadmed has joined #asahi-dev
01:18EnnuiLHi, I have tried to tackle the software Fn key issue years ago and well, I was wondering if this issue is still something I could try to tackle; my conclusion back then has been that progress was blocked on libxkbcommon and friends and well, things have changed since then, but now I'm suspecting that there's not much that can be done..? At the very least I have ruled out SDL support for an Fn modifier
02:02dvh has joined #asahi-dev
02:23linkmauve has quit [Remote host closed the connection]
02:28linkmauve has joined #asahi-dev
03:00yurix has quit [Remote host closed the connection]
03:00yurix has joined #asahi-dev
03:05tobhe_ has joined #asahi-dev
03:12tobhe has quit [Ping timeout: 480 seconds]
03:44linuxgemini has quit [Quit: getting (hopefully fresh) air o/]
03:48linuxgemini has joined #asahi-dev
04:16nora_ has joined #asahi-dev
04:17nora has quit [Ping timeout: 480 seconds]
04:39umo is now known as Guest302
04:39umo has joined #asahi-dev
04:42umo has quit [Remote host closed the connection]
04:42umo has joined #asahi-dev
04:47Guest302 has quit [Ping timeout: 480 seconds]
05:45umo has quit [Remote host closed the connection]
05:45umo has joined #asahi-dev
06:02Guest4545 has quit [Read error: Connection reset by peer]
06:45bit_runner80 has joined #asahi-dev
06:58procee has joined #asahi-dev
07:08Calandracas_ has joined #asahi-dev
07:14Calandracas__ has joined #asahi-dev
07:14Calandracas has quit [Ping timeout: 480 seconds]
07:19Calandracas_ has quit [Ping timeout: 480 seconds]
07:32Calandracas_ has joined #asahi-dev
07:32bit_runner80 has quit [Read error: Connection reset by peer]
07:38Calandracas__ has quit [Ping timeout: 480 seconds]
07:42yukaI should borrow my sister's M5 and check if I can get USB running on it given that I have an SPMI gen 4 driver now
07:56marcusr has quit [Remote host closed the connection]
07:59marcusr has joined #asahi-dev
08:18ravikant_ has joined #asahi-dev
08:25Bhyve has joined #asahi-dev
08:32___nick___ has joined #asahi-dev
08:33elara_ has joined #asahi-dev
08:34elara has quit [Ping timeout: 480 seconds]
08:34elara_ is now known as elara
08:37chadmed has quit [Remote host closed the connection]
08:37chadmed has joined #asahi-dev
08:40chadmed has quit []
08:40chadmed has joined #asahi-dev
08:53snooze6214[m]sven: I saw that m5 hv got merged, does that include sk emulation? I can try on the a18 pro tonight :)
08:54snooze6214[m]Should I make a PR for that hack? Or is it too "hacky" to be in upstream m1n1?
08:54svenhttps://asahilinux.org/docs/sw/m1n1-hypervisor/#running-the-hypervisor-with-sptmtxmcl4-on-m5
08:55elara has quit [Ping timeout: 480 seconds]
08:55snooze6214[m]Cool, thanks :)
09:35N_Tempus_04 has joined #asahi-dev
10:22chaos_princessCan someone with a m5 pro/max check if SYS_IMP_APL_VM_TMR_FIQ_ENA_EL2 works
10:26yukamy fingerbot setup sadly is not reliable yet, so it got the M5 Pro stuck in boot selection :/
10:34elara has joined #asahi-dev
11:15Bhyve has quit []
11:53N_Tempus_04 has quit [Ping timeout: 480 seconds]
12:13floklijannau: myria opened https://github.com/AsahiLinux/linux/issues/673 about the build regression with tyr I wrote about earlier.
12:27myriajannau: in response to the comment, we just use the default nixos kernel config with asahi toggles, which includes all DRM devices...
12:37chaos_princessif you want to test stuff on a rk3588 - i have one and can help
12:37jannauprobably not for the asahi specific kernel but the default nixos kernel probably has some
12:39jannaunah, I'll try to remember to ask on thursday and point the tyr devs to the nixos kernel which already "ships" tyr
12:42ravikant_ has quit [Ping timeout: 480 seconds]
12:47jcbm has quit [Ping timeout: 480 seconds]
12:52jcbm has joined #asahi-dev
12:59ClmentVerna[m] has left #asahi-dev [User left]
13:00ravikant_ has joined #asahi-dev
13:17cy8aer has quit [Remote host closed the connection]
13:23cy8aer has joined #asahi-dev
13:36Bhyve has joined #asahi-dev
13:40picnoir has quit [Quit: WeeChat 4.10.2]
13:41picnoir has joined #asahi-dev
13:42N_Tempus_04 has joined #asahi-dev
13:59steven has quit [Remote host closed the connection]
14:00flokliBut if there's some consensus on not building this by default, it should probably not be in the distro kernel
14:04AnEagle[m] has joined #asahi-dev
14:04AnEagle[m]Hi, how much space should I account for to install the m1n1 hv ?
14:05chaos_princessabout 1 macos install
14:06AnEagle[m]Are we talking the 80gb that the asahi installer keeps, or the ~30gb of macos data
14:08chaos_princessdepending on how you are installing the macos you intend to overwrite with m1n1. You can either do it in an existing container, where they share space, or dedicated partition.
14:29dvh_ has joined #asahi-dev
14:30AnEagle[m]Thanks, I'll see what I can figure out
14:34dvh has quit [Ping timeout: 480 seconds]
14:37dvh_ is now known as dvh
14:38vaibhavxlr has quit [Quit: Connection closed for inactivity]
14:42N_Tempus_04 has quit [Ping timeout: 480 seconds]
14:49chadmed has quit [Quit: Konversation terminated!]
14:50chadmed has joined #asahi-dev
14:53zaprit has joined #asahi-dev
14:57boubou has joined #asahi-dev
14:58boubou is now known as Guest342
14:59Guest342 has quit []
14:59agentboubou has joined #asahi-dev
15:08zaprit has quit [Remote host closed the connection]
15:09zaprit has joined #asahi-dev
15:10zaprit has quit [Remote host closed the connection]
15:11zaprit has joined #asahi-dev
15:13N_Tempus_04 has joined #asahi-dev
15:16Yushi has quit [Remote host closed the connection]
15:17Yushi has joined #asahi-dev
15:18Bhyve has quit []
15:29zaprit has quit [Quit: Konversation terminated!]
15:33zaprit has joined #asahi-dev
15:45Yushi has quit [Remote host closed the connection]
15:46Yushi has joined #asahi-dev
15:47okb_ has joined #asahi-dev
15:49okb__ has joined #asahi-dev
15:49okb has quit [Ping timeout: 480 seconds]
15:51Yushi has quit [Remote host closed the connection]
15:52Yushi has joined #asahi-dev
15:54__nick__ has joined #asahi-dev
15:54chadmed has quit [Read error: Connection reset by peer]
15:54chadmed has joined #asahi-dev
15:56okb_ has quit [Ping timeout: 480 seconds]
15:57___nick___ has quit [Ping timeout: 480 seconds]
16:03chadmed has quit [Ping timeout: 480 seconds]
16:49thelounge55 has joined #asahi-dev
16:49thelounge55 has quit []
16:50thelounge55 has joined #asahi-dev
16:59okb__ has quit [Ping timeout: 480 seconds]
17:00hardy94 has joined #asahi-dev
17:22___nick___ has joined #asahi-dev
17:25__nick__ has quit [Ping timeout: 480 seconds]
17:29ravikant_ has quit []
17:34Bhyve has joined #asahi-dev
17:44N_Tempus_04 has quit [Ping timeout: 480 seconds]
17:45zaprit has quit [Quit: Konversation terminated!]
17:49___nick___ has quit [Remote host closed the connection]
17:51snooze6214[m]sven: hv on a18 pro: it seems to get stuck in a panic loop pretty early in the boot: https://termbin.com/0a3mn... (full message at <https://matrix.org/oftc/media/v1/media/download/Ae0D_OvRy_lizPWSYpfcuKelH4rN8Me8SrKAFe6Z8Agobyo4RT10A7rFzLU4VZv2NnBb4zcXY--RxuKphAeFdbxCeiCYkdYwAG1hdHJpeC5vcmcvUlFKYktXQkVZanFkbURQYkdwbGJxUnpY>)
17:53snooze6214[m]oh okay :0 i didn't see the early "Darwin" log in XNU logs, so i thought it was early in the boot
17:54svenI think cL4 is running and it’s failing somewhere while the exclaves root task is initialized
17:55svenDERComponentGraph Sounds like it’s trying to parse some der certificate file, maybe my heuristic for trustcache and integrity-catalog is wrong? they might contain just that
17:56snooze6214[m]26.6.2
17:57snooze6214[m]should i be trying something more recent?
17:57svenreadmem doesn’t work because you’d need to translate that PC via cL4’s pagetable to a physical address first
17:58snooze6214[m]ah, so ill have to walk the pagetables manually and translate?
18:00snooze6214[m]yeah, i mean there was only one pair i think
18:03rivendell has quit [Remote host closed the connection]
18:04rivendell has joined #asahi-dev
18:07rivendell has quit [Remote host closed the connection]
18:07rivendell has joined #asahi-dev
18:15zaprit has joined #asahi-dev
18:27okb has joined #asahi-dev
18:48okb has quit [Quit: Leaving]
18:48rivendell has quit [Remote host closed the connection]
18:50rivendell has joined #asahi-dev
18:51Bhyve has quit []
19:00bgb has quit [Remote host closed the connection]
19:00bgb has joined #asahi-dev
19:06zaprit has quit [Quit: Konversation terminated!]
19:07svensnooze6214[m]: just to confirm: the files you extracted are from the same macos version that's installed?
19:07fake_agent_smith has joined #asahi-dev
19:09zaprit has joined #asahi-dev
19:13svenhrm.. so running strings on that bundle shows a lot of debug crap but i've seen any of that on serial
19:14Deewiant has quit [Remote host closed the connection]
19:15Deewiant has joined #asahi-dev
19:24sveni guess it somehow has to pass that to XNU anyway so we probably wouldn't get anything this early either way :(
19:26fake_agent_smithhi, letting you know just in case, fairydust applies cleanly on top of asahi-7.3.3-1 and I haven't noticed any issues on my j414s
19:28vimproved has quit [Remote host closed the connection]
19:30vimproved has joined #asahi-dev
19:38dvh has quit [Remote host closed the connection]
19:40dvh has joined #asahi-dev
19:41cy8aer has quit [Remote host closed the connection]
19:42snooze6214[m]sven: sorry i was afk. yes, i do have serial=3 in my boot args, removing that actually does take me further :0
19:42snooze6214[m]like i see the login screen lol \o/
19:43snooze6214[m]not sure, let me try again
19:44chaos_princessit was a thing at least at 13.5 that enabling every single pmgr log made it fail to boot as the logging code took mutexes from an interrupt context or sth
19:44chaos_princessprobably the same, serial=3 needing a mutex, dockchannel - not so much
19:47snooze6214[m]nvm not consistant
19:47snooze6214[m]im stuck again :/
19:48bgb has quit [Remote host closed the connection]
19:52cy8aer has joined #asahi-dev
19:54bgb has joined #asahi-dev
19:55snooze6214[m]nvm sorry i feed dumb, i had forgotten i had booted into my macos install to make sure i fetched the right files. then i went afk and forgot about it
19:55snooze6214[m]i consistantly get stuck on the panic loop i think
19:57sventhat would've been very useful because it might just tell us what it hates before that panic
19:57snooze6214[m]exclave serial out as in? all my output comes from a printf i added above handle_vuart :')
19:58snooze6214[m]i got all excited for no reason -_-
19:58svenif you run strings on the exclavebundle there's a lot of stuff that looks like very verbose output
20:00svenI suspect it doesn't like something in the device tree because those were the early panic I got and they were somewhat similar
20:00snooze6214[m]$ strings exclavecorebundle.b | grep 'try!'... (full message at <https://matrix.org/oftc/media/v1/media/download/AYVFpRIDMKdghZWAFNvncJiCgzPau1vB5-OLNGIjWtGfdVh7a9PAAEX77u-NSwiA8F_Pn_BQ_-sd_DxsyhCHsv9CeiCf7I5QAG1hdHJpeC5vcmcvSktZaVh6S3ZLbmxHdEpza2NqamJtVWVm>)
20:01svenyes.. it's even more annoying because it does have that DERComponentGraph.swift string but I think the panic is along the lines of "... %s:%d", "DERComponentGraph.swift", 157
20:01svenso I'd have to reverse engineer where it puts 157 into some register to figure out what's going on
20:01snooze6214[m]will p.hv_translate give the address on host that i can prod with iface.readmem?
20:02snooze6214[m]im learning a lot about arm just by doing this lol xD
20:02snooze6214[m]but i have no clue what im doing tbh
20:09snooze6214[m]>>> p.hv_translate(0x0000c005b4c4)
20:09snooze6214[m]no luck ig :/
20:10snooze6214[m]* ```... (full message at <https://matrix.org/oftc/media/v1/media/download/AWt6y9mIZjCPFzkjHUbhVHH0H_B7efK79k-lUBmuzp2kuJimFafUbYEQ1hvHpnHEcbZvgslL1l3APhKOzxq1jXNCeiCgefQgAG1hdHJpeC5vcmcvWklvQ3NBWUlDR05MQVd2THlwRUdBb0FV>)
20:10snooze6214[m]* ```... (full message at <https://matrix.org/oftc/media/v1/media/download/Abf-nBEu3rCNhrCdbPkO2AdIt3mQzUjQN10sfFRYXM2As4-nGIUYCCX8z3pkhKY_hOmcijGc_90WCMR8lnTZhVpCeiCgfHVAAG1hdHJpeC5vcmcvWEFOak1BZ1Z3cmxGYWxSUWJmdUdEc2pF>)
20:10snooze6214[m]sven: i can try 157 in some register. let me check
20:11snooze6214[m]* can try finding 157 in
20:11snooze6214[m]* can try finding 157 in, * some register/addresses pointed by registers. let
20:13dvh has quit [Remote host closed the connection]
20:13dvh has joined #asahi-dev
20:21snooze6214[m]also, may i ask when it is okay to disassemble apple binaries? so far i haven't (other than the disassembly in ctx() m1n1 output)
20:22snooze6214[m]ah okay, cool. makes sense
20:23snooze6214[m]yeah makes sense :')
20:25sveni'd try patching that mov w4, #0x9d in the binary to something that escapes to the hv shell
20:26snooze6214[m]like a hvc call of some sort?
20:45sventry self.adt.create_node("/chosen/exclave-misc").bootargstring = self.tba.cmdline in proxyclient/m1n1/hv/__init__.py after e.g. self.adt["product"].exclaves_enabled = 1
20:46chaos_princesssven: we had hv calls in m1n1 for gpu stuff, no?
20:46svenbut i trap all exception to EL1 anyway so something like svc #whatever would work i guess but that's annoying
20:47svenbut I think I found what's wrong anyway. or, well, i have an educated guess because I didn't follow the entire path from failure to that exception
20:50snooze6214[m]sven: holy fuck this worked!!!
20:51sveni dunno why that isn't fatal on m5, it just complains if the exclave memory map is missing there
20:53mmediouni[m] has joined #asahi-dev
20:53mmediouni[m]sven: a trick to go EL0 to EL2 (if not VHE in-host) is a page mapped at stage 1 but not at stage 2
20:53svenmmediouni[m]: yeah, that's what I already do in the SPRR/GXF emu but hooking that up and exiting to the hv shell sounds annoying
20:55svenoh wait.. you mean map a page to EL0 that has not stage 2 backing and then try to read from that one?
20:56mmediouni[m]Or use MDCR_EL2.TDE and do a breakpoint instr
20:56chaos_princesssven: ok, so, we are not using hvc, we are using brk and mdcr_el2.tde
20:58* snooze6214[m]sent a code block: https://matrix.org/oftc/media/v1/media/download/AYH-sht2rnxT8HGcwNuRER2PvlC9Gz_Seug6nJjpeBLLY_m-w9PlKSYr8wPJ3IKX6-rkFy0DSt6-8JE0ZC6KT8JCeiCjOzsQAG1hdHJpeC5vcmcvWlBZWmJoaGxIeE5uR3JHeWtSd2JzT1hX
20:59mmediouni[m]Watchtower back in the pre-KTRR days did a dance where it either trapped neon execution to EL3 or trapped interrupts to be somewhat guaranteed to run constantly without much of a perf impact
21:01mmediouni[m]But EL3 just didn't have enough at the time to have a good story to protect the kernel - EL3 FGT came much later
21:01mmediouni[m]so apple switched to KTRR
21:03mmediouni[m]I think there's a boot arg to disable the hw wkdm path
21:04snooze6214[m]okay, ill just memcpy those pages, lemme try :)
21:04snooze6214[m]there's a bootarg to disable this?
21:05agentboubouHeya, new here but I have a working gpu driver for t6030 (m3 pro). I've used AI during the development and I know this is something Asahi doesn't approve of. Is this something that the team would still want to take a look at ? I got 120hz + vulkan working. It's far from perfect, but it's pretty smooth
21:10chaos_princesstry vm_compressor_codec=1
21:11zaprit has quit [Quit: Konversation terminated!]
21:11mmediouni[m]wksw_force
21:11mmediouni[m]vm_compressor_force_sw_wkdm
21:13mmediouni[m]Fun errata with retries (!) in OSS XNU code
21:13mmediouni[m]https://github.com/apple-oss-distributions/xnu/blob/f6217f891ac0bb64f3d375211650a4c1ff8ca1ea/bsd/dev/arm64/sysctl.c#L331
21:18snooze6214[m]chaos_princess: this makes the hv restart as soon as the secondary cpus start up for some reason :(
21:18chaos_princessdid they come up before?
21:18chaos_princessif so, use the other two flags
21:19snooze6214[m]infact, took me till the login screen before, and then hung
21:21mrkajetanp has joined #asahi-dev
21:21mrkajetanp_ has quit [Read error: Connection reset by peer]
21:22evrin has joined #asahi-dev
21:23chaos_princesshow much memory is m1n1 stealing for the page tables on the neo?
21:25chaos_princessthat's out of 8
21:27snooze6214[m]hmm so "vm_compressor_codec=1" and "wksw_force=1" both give me an exception on when secondary CPUs come up. "vm_compressor_force_sw_wkdm=1" hangs here:... (full message at <https://matrix.org/oftc/media/v1/media/download/AUo1dYuDO1GH2vCSAad1SC6TatMYn8QckABe9YeS_-Jq-HvEHdyAPt5VlQrmnHuBs6eeF8oh2MI3S-bTBEYPa-lCeiCk7SyQAG1hdHJpeC5vcmcvWFRIVlR3V2hwSnh1UmZlVEhpZFdPZ0Rw>)
21:28iwuvmika has joined #asahi-dev
21:32zaprit has joined #asahi-dev
21:32chaos_princessdo you have symbols
21:33snooze6214[m]no i dont, sorry :/
21:33snooze6214[m]ill look into setting up a kdk kernel tomorrow... too late today :') (3 am-ish for me)
21:34chaos_princesslike, why is it trying to hvc, that is not a hv kernel
21:34chaos_princesscan you disasm a larger block
21:34snooze6214[m]yeah, wait
21:34yukaI would be interested in how to get symbols in 26/27, wasn't able I haven't figured it out. ried the release kernel with the accompanying dSYM DWARF passed as -s but just doesn't give me any symbols.
21:37snooze6214[m]ye sure :)
21:51snooze6214[m]pardon my language, but fuck me ig.... i tried "vm_compressor_force_sw_wkdm=1" again it got me to the login screen with the same panic... (full message at <https://matrix.org/oftc/media/v1/media/download/Ae9oNn9Nj7DM19iXpYwtYH9SSNhJCjukLapGWVwN5nxOAGERqHPaxL24lxmoL6TjdFYAL7lM1NouGGZWKf_-6CNCeiCmSJBwAG1hdHJpeC5vcmcvVmNrUlZtaXRWRXFNRUhrTHljUmhsWWRk>)
21:51snooze6214[m]i should probably try a few times before i post here huh :'
21:51snooze6214[m]* huh :')
21:53snooze6214[m]python3 proxyclient/tools/run_guest.py --sptm-mode=emulation --sptm ../kernelcache/neo_dump/sptm.macho --txm ../kernelcache/neo_dump/txm.macho --exclave-bundle ../kernelcache/neo_dump/exclavecorebundle.b --exclave-... (full message at <https://matrix.org/oftc/media/v1/media/download/ARFicHRjs_VKPrUK77PQZBLdEePzeNDiBxNXZNKLg6Morvaz1WUxcRecuwSZTCAH4bhSL_uvH9Cx53P6AU2MAbxCeiCmYPVwAG1hdHJpeC5vcmcvWmRkV0tOZnlHeFNrRWFlT1ZEdk5OU1By>)
21:54snooze6214[m]sven: yep, i always do chainload a fresh build of m1n1 before i run_guest :)
21:55svenif the bootarg doesn't work for some reason you'll have to patch wkdmc/wkdmd like i e.g. patch genter or that weird GXF TLBI instruction
21:56svenmemcpy a page (16K? who knows) from rS to rD and you'd have to figure out the return status in r5 if https://asahilinux.org/docs/hw/cpu/apple-instructions/ is correct
21:56snooze6214[m]makes sense, ill just have to patch it and in the hv call, probably just memcpy?
21:56snooze6214[m]gotcha, i can try that '
21:56snooze6214[m]* try that :', * ')
21:57chaos_princessxnu has the spicy instructions censored
21:58iwuvmika_ has joined #asahi-dev
21:59mmediouni[m]KernelSupport alone is such a travesty unfortunately
21:59chaos_princesscan you disasm like 20 instructions back from the wkdm
22:03iwuvmika has quit [Ping timeout: 480 seconds]
22:04snooze6214[m]sven: ayooo this worked! \o/
22:05snooze6214[m]god bless xD
22:05N_Tempus_04 has joined #asahi-dev
22:05chaos_princessi was about to pick a cbnz and turn it into a cbz :P
22:06svenso now you can create three commits an open a PR: that CTRR alias hack, that weird exclave-misc ADT thing, and finally add that bootarg automatically. there's already a list somewhere in hv/__init__.py that disables a bunch of other crap
22:06evrin has quit [Quit: evrin]
22:07* snooze6214[m]uploaded an image: (260KiB) < https://matrix.org/oftc/media/v1/media/download/Ad-jnxmlX5bDxoKw2HPA-F49-QvKiM6xGA8zKp_H_kVszyEd5uN5kqaqxs2cO3PTAqXI12uGpNtdF3Mn2kJT3qNCeiCnOXAwAG1hdHJpeC5vcmcvWk9TUHZyckhmT0FMVHpmVlRUWENnQVpm >
22:08chaos_princesshell yes, congrats
22:08chaos_princessnow time to figure out the cursed usb switch?
22:09snooze6214[m]sven: Ack makes sense. Will do in a bit :)
22:09snooze6214[m]chaos_princess: Yes, although it might take a while with my constant skill-issue lol
22:11pipi has joined #asahi-dev
22:14N_Tempus_04nice
22:17pipi has quit [Quit: Page closed]
22:29zaprit has quit [Ping timeout: 480 seconds]
22:31N_Tempus_04 has quit [Ping timeout: 480 seconds]
22:34bgb has quit [Remote host closed the connection]
22:35ayaan` has joined #asahi-dev
22:38bgb has joined #asahi-dev
22:49ayaan` has quit [Ping timeout: 480 seconds]
22:58ayaan` has joined #asahi-dev
23:00truita has quit [Remote host closed the connection]
23:00barusu has quit [Remote host closed the connection]
23:00barusu has joined #asahi-dev
23:00truita has joined #asahi-dev
23:05ayaan` has quit [Remote host closed the connection]
23:06ayaan` has joined #asahi-dev
23:22ayaan` has quit [Ping timeout: 480 seconds]
23:29ayaan` has joined #asahi-dev
23:38ayaan` has quit [Remote host closed the connection]
23:38ayaan` has joined #asahi-dev
23:40ayaan` has quit [Read error: Connection reset by peer]
23:59hardy1994 has joined #asahi-dev
23:59hardy94 has quit [Read error: Connection reset by peer]
23:59hardy1994 is now known as hardy94