2026-05-05
00:10enick_22711 has joined #asahi-dev
00:10rhysmdnz111 has joined #asahi-dev
00:15rhysmdnz11 has quit [Ping timeout: 480 seconds]
00:16enick_2271 has quit [Ping timeout: 480 seconds]
00:19ddxtanx has joined #asahi-dev
00:34hardy94 has joined #asahi-dev
01:10pb17 has quit [Ping timeout: 480 seconds]
01:10pb17 has joined #asahi-dev
01:30vols has joined #asahi-dev
01:31testuser has joined #asahi-dev
01:31testuser has quit [Remote host closed the connection]
01:33testuser has joined #asahi-dev
01:44testuser has quit [Remote host closed the connection]
01:50hardy94 has quit [Ping timeout: 480 seconds]
01:51cow123 has quit [Remote host closed the connection]
01:51cow123 has joined #asahi-dev
01:56testuser has joined #asahi-dev
02:16tobhe has joined #asahi-dev
02:19testuser has quit [Remote host closed the connection]
02:19testuser has joined #asahi-dev
02:23tobhe_ has quit [Ping timeout: 480 seconds]
02:43testuser has quit [Remote host closed the connection]
03:04hexdump01 has joined #asahi-dev
03:05hexdump0815 has quit [Ping timeout: 480 seconds]
03:11troseman has joined #asahi-dev
03:12troseman has quit [Remote host closed the connection]
03:13troseman has joined #asahi-dev
03:18testuser has joined #asahi-dev
03:41voidah has quit [Remote host closed the connection]
03:42voidah has joined #asahi-dev
03:44Stary has quit [Quit: ZNC - http://znc.in]
03:50Stary has joined #asahi-dev
03:51dataviruset has joined #asahi-dev
03:54nora has joined #asahi-dev
03:56nora_ has quit [Ping timeout: 480 seconds]
04:05r0ni has joined #asahi-dev
04:08testuser has quit [Ping timeout: 480 seconds]
04:35testuser has joined #asahi-dev
05:06lukeleppan9 has quit [Read error: Connection reset by peer]
05:06lukeleppan9 has joined #asahi-dev
05:08lukeleppan9 has quit []
05:20lukeleppan9 has joined #asahi-dev
05:32ddxtanx has quit [Quit: Konversation terminated!]
05:35lukeleppan9 has quit [Quit: Ping timeout (120 seconds)]
05:38lukeleppan9 has joined #asahi-dev
05:44NSA_RECON_BOT has joined #asahi-dev
05:45NSA_RECON_BOT Hi, I am reverse engineering drivers but I cannot get the Asahi installer to give me the option to change the m1n1 to proxy mode, no matter what I do, it does not go into proxy mode. I have dual boot Asahi / MacOS 13.5 and need to get MacOS running atop of this hv so that I may dump out MMIO accesses to my other Mac. Any help would be appreciated, cheers.
05:48JamesCalligeros[m]You need to either set up a second macos and replace the kernelcache with m1n1 using kmutil, or do it to your main macos install. You cannot use the asahi boot container for macos hypervising
05:50unifiedgravity[m] has joined #asahi-dev
05:51unifiedgravity[m]Your irc bridge must be broken
05:52unifiedgravity[m]Yeah, I am trying to do it to the main MacOS install, can this be done with the asahi installer or do I have to do it manually?
05:52JamesCalligeros[m]You have to do it manually. Read the hypervisor setup docs
05:52unifiedgravity[m]I mean the hv output on the boot screen now references USB ports but connection not detected.
05:52unifiedgravity[m]Even though it's definitely connected to the other Mac
05:53unifiedgravity[m]If you are referring to https://asahilinux.org/docs/sw/m1n1-hypervisor/, nothing in that seems to be about what I am trying to do
05:53JamesCalligeros[m]What are you trying to do that is different? That doc is the only supported approach to virtualising macos for RE.
05:54kaazoo has quit [Quit: Leaving.]
05:54unifiedgravity[m]"Preparing the macOS Volume by disabling security features" seems to be the only relevant part, and I thought I've already done all that.
05:54JamesCalligeros[m]Did you replace the macos kernel of the container youre trying to virtualise with m1n1?
05:55berzrk_ has joined #asahi-dev
05:55JamesCalligeros[m]Did you extract the kernelcache to upload to the DUT using the proxyclientv
05:57unifiedgravity[m]No, I am not interested in uploading kernels. I just want m1n1 to enter proxy, boot the kernel that came with the MacOS installation. Am I trying to do something that does not make sense here?
05:57JamesCalligeros[m]Yes. You need to do it the way the documentation says. There is literally no other way to do it
05:57JamesCalligeros[m]You cannot just magically have the DUT chainload a kernel and automatically connect to the host machine
05:57lukeleppan9 has quit [Quit: Ping timeout (120 seconds)]
05:58JamesCalligeros[m]The host machine uploads the kernel and tells the DUT to jump to it
05:58berzrk has quit [Ping timeout: 480 seconds]
05:58unifiedgravity[m]Okay I'll try to understand it. If I sort it out then I will amend the documentation to make it clearer in ways that would have helped me and you can consider updating the docs if my edits make sense.
05:59unifiedgravity[m]But I just can't get the USB serial device to appear on the host machine from the target machine
05:59lukeleppan9 has joined #asahi-dev
06:01Pokemod977362084712818 has quit []
06:03mildsunrise has quit [Quit: disconnect]
06:04unifiedgravity[m]Does it matter if I use a C-C 2.0 cable, USB4 cable or a USB-C to USB-A to USB-C with target as peripheral/downstream
06:04mildsunrise has joined #asahi-dev
06:05Pokemod977362084712818 has joined #asahi-dev
06:08JamesCalligeros[m]Active Thunderbolt/USB4 cables may not work. The stock apple charging cable does though
06:08lukeleppan9 has quit [Quit: Ping timeout (120 seconds)]
06:08unifiedgravity[m]Hm, I just booted MacOS and it did m1n1 but flew past, never paused for proxy, and this never changed
06:08unifiedgravity[m]$ ./tools/picocom-sec.sh
06:08unifiedgravity[m]Waiting for UART device file '/dev/cu.usbmodemP_03' to appear (Ctrl+C to abort)...
06:10unifiedgravity[m]I have a couple of iPhone 60W cables somewhere but I can't find them for the life of me.
06:10JamesCalligeros[m]Do what the documentation says. Replace whatever your container’s boot object currently is with m1n1 and only m1n1. If there is nothing concatenated on the end of m1n1, it will start the proxy and wait for instructions from the host.
06:11unifiedgravity[m]Okay thanks, jumping into the MacOS partition's 1tr, will see
06:14testuser has quit [Remote host closed the connection]
06:15unifiedgravity[m]Yeah, I will propose amendments to the docs, whilst they may be technically sufficient, a heap of notes for instance to note that the "recovery" option in boot options goes to the recovery image of the currently selected default boot disk, that there is a 1tr for every boot option and they are not the same. And to disable Filevault etc etc. A note that the kernel must be uploaded, it cannot boot the one from the partition, etc.
06:15unifiedgravity[m]These things would make it easier.
06:20unifiedgravity[m]kmutil configure-boot \... (full message at <https://matrix.org/oftc/media/v1/media/download/AZxBc9EIxDqJ2uvj_agTrDKVJ14PrpRxWe9Mfx2acdnH3Lmx3VjJkm8A6bY7mz9qaohORwyv8HmjT8zn2UAqk9BCeePuSjQwAG1hdHJpeC5vcmcvWEFGTmVpUU5pVnByUmxNekpmTmVYenh0>)
06:21unifiedgravity[m]And besides even if I mounted it, when it booted using that filepath, it wouldn't necessarily be the same path when in the bootloader
06:22JamesCalligeros[m]Serve the build from your host and copy it to 1tr with curl or whatever
06:22unifiedgravity[m]Ah thanks so I take it that it writes the bin file somewhere it can find it later
06:26unifiedgravity[m]Anyways my practice will be to finish the Linux Presonus 2626 driver first (it is structurally there, but seems to be missing something in the init sequence) but then I will be doing the USB4 host bringup for Asahi.
06:29svenhttps://github.com/AsahiLinux/linux/commits/sven/tbt-wip/ that's 95% of tbt bringup, there's "just" some bug in there that prevents the NHI to see the connection
06:30NSA_RECON_BOTOh damn, that's impressive, I had no idea there was that branch
06:32unifiedgravity[m]Do you have any idea of the feasibility of porting lspci to MacOS? It will be interesting to finally see the whole PCIe layout when Linux can bring up the domains for USB4 PCIe. I have no way of visualising it easily in MacOS. ioreg seems like a blunt mallet
06:33vols has quit []
06:33unifiedgravity[m]No, it won't, but it will be interesting
06:33unifiedgravity[m]Cool, I will see what I can do with it
06:34unifiedgravity[m]Anyways brew doesn't like my host config so it's compiling llvm from source so I'll have to wait for that so I can build m1n1 and then I will let you know what happens. Cheers
06:37unifiedgravity[m]I saw something a few months ago about M3 running but the progress pages hadn't been updated to reflect so I guess it's still alpha/WIP but that will be awesome too. I got an M2 Air 15 as a sacrifice target. Will be excited to help with platform drivers for M3-M5 once bring-up is completed. I don't think I am capable of helping for the initial platform bring-up work.
06:38lukeleppan9 has joined #asahi-dev
06:39unifiedgravity[m]To clarify, the post which was made a few months ago about M3 running. I saw it about a week ago.
06:40JamesCalligeros[m]I’d describe M3 as “bootable”. Storage, kb, trackpad, usb and display via firmware framebuffer work but it’s not a pleasant daily driving experience
06:42unifiedgravity[m]I don't expect it to be pleasant; my understanding of this project is to get the kernel changes mainlined and that Asahi is for development, and then any repo can then support the platform in stable releases.
06:42unifiedgravity[m]So if I buy a M3, will the Asahi installer work?
06:43chaos_princessif you want an install on m3, go manual route or patch the installer
06:44unifiedgravity[m]By the way, search Ebay for "Macbook b-grade" you can find Macs with some scratches which don't affect anything but knock the price down a lot, occasionally. I got my Macbook Air 15 M2 16GB/1TB for equivalent of about $786 USD.
06:44unifiedgravity[m]I can patch the installer I guess, cheers
06:46pb17 has quit [Ping timeout: 480 seconds]
06:49mripard has joined #asahi-dev
06:50mripard_ has quit [Ping timeout: 480 seconds]
06:55voidah has quit [Remote host closed the connection]
06:56voidah has joined #asahi-dev
06:57pb17 has joined #asahi-dev
07:01c10l has quit [Quit: Ping timeout (120 seconds)]
07:06jannauunifiedgravity[m]: see https://github.com/jannau/asahi-installer/tree/wip-m3-ipsw-14.6.1 for the patched installer
07:13Felix[m]1I guess when people talk about the M3 here, they’re referring only to the base model, which does not include M3 Max?
07:14lukeleppan9 has quit [Quit: Ping timeout (120 seconds)]
07:14jannauthe fedora44 images from the installer are not bootable because m1n1 and u-boot miss m3 support
07:14lukeleppan9 has joined #asahi-dev
07:15jannauFelix[m]1: generally yes, M3 Pro/Max/Ultra support is behind but I'm planning on bringing it to the same lavel in thext few days
07:25lukeleppan9 has quit [Quit: Ping timeout (120 seconds)]
07:25lukeleppan9 has joined #asahi-dev
07:28unifiedgravity[m]Sorry seem to have botched the Asahi boot, how can I get the installer to repair the bootloader for Asahi?
07:28chaos_princessjust kmutil it by hand, it is easier
07:29unifiedgravity[m]It just shows the Asahi logo, no key presses make it show the error or dmesg so I have no idea what's wrong. I tried kmutil and idk how to get it back to normal
07:29chaos_princessthat is how it is supposed to work, m1n1 does not take input
07:30chaos_princessuse serial and a second machine to drive it
07:30unifiedgravity[m]Like what if the boot efi partitions are broken kmutil will not fix that
07:30chaos_princessm1n1 does not know anything about EFI, that is u-boot's job
07:30unifiedgravity[m]As I have been saying, m1n1 refuses to enter proxy mode or show a USB serial to another machine
07:30chaos_princessare you using release m1n1 from installer or custom build?
07:31unifiedgravity[m]Just release
07:31chaos_princessboot to recoveryos, do `csrutil disable` on m1n1 container
07:31chaos_princessand i guess `nvram boot-args="-v"`
07:32unifiedgravity[m]Because I can't get the installer into repair mode I can't reinstall m1n1 in proxy mode
07:32unifiedgravity[m]It's always been disabled and the boot args always -v
07:32chaos_princesson linux container?
07:32chaos_princessare you sure
07:32chaos_princessSIP state is per-install
07:33unifiedgravity[m]Wait so there's a 1tr for Linux and a 1tr for MacOS but you're saying there's a third recovery for m1n1?
07:34chaos_princessrun `diskutil list` from macos, paste here
07:34chaos_princess(via a pastebin ofc)
07:34unifiedgravity[m]One sec
07:39chaos_princessyou should be targeting disk4 with csrutil
07:41unifiedgravity[m]Yeah I was just in the 1tr for that partition (I put Linux as the boot disk then went into options in boot options)
07:41chaos_princessrun `csrutil status` from recoveryos, pastebin the result
07:41unifiedgravity[m]This doesn't stop the Asahi from hanging on its logo when I try to boot it
07:42unifiedgravity[m]I need to find a way to get the installer to wipe the Asahi's boot partitions / m1n1 etc and reinstall them because something must have been corrupted
07:44chaos_princessthe first stage m1n1 is healthy, as you are reaching the boot logo at all. (if it was corrupted, iboot will not start it at all and kick you to recovery)
07:44unifiedgravity[m]Anyways I dumped it to a file, rebooting intomacos to post
07:45unifiedgravity[m]Okay but it's not getting to grub anymore
07:47chaos_princesscsrutil disable the linux container again
07:50unifiedgravity[m]Interesting. Still hangs but this time it said waiting for proxy connection for the first time ever. But yeah still does not boot after the proxy timeout
07:50chaos_princesshow exactly does it "not boot"
07:51unifiedgravity[m]After that proxy timeout it shows the Asahi logo and then nothing else happens and can't get it to show logs
07:51chaos_princessso it clears the log?
07:52unifiedgravity[m]I don't know about clearing but the log that shows proxy timeout disappears and all that is left is the Asahi logo
07:52chaos_princessok, it booted stage 2, which is what is hanging
07:54chaos_princessgo mount ESP from macos, there should be a m1n1/boot.bin file on it, and a boot.bin.old next to it. Replace the .bin with .old
07:56jannauif boot.bin.old doesn't work, you can try the one from https://github.com/AsahiLinux/m1n1/actions/runs/25250750280
07:58unifiedgravity[m]Nice thanks booted to GRUB again
07:58unifiedgravity[m]I wonder what happened to boot.bin
07:58unifiedgravity[m]Anyhow
07:59unifiedgravity[m](venv) nicholas@nicholas-mac:~/git/asahi/m1n1/proxyclient$ while true; do ls /dev/cu.usbmodem* 2>/dev/null; sleep 0.5; done... (full message at <https://matrix.org/oftc/media/v1/media/download/AZ1CjirUThLL1ZUAbOgg9f7xGpr-rcyQC0XNaPiy_UA7AuQw33rPfp9O-eCok4NZ2mS8T0oQDQAqbMDcKGrEFXBCeePz-TqgAG1hdHJpeC5vcmcvelJLb3ZUclJsY1daQUZvY3ZlRkZqVFFU>)
07:59unifiedgravity[m]s/(venv) nicholas@nicholas-mac:~/git/asahi/m1n1/proxyclient//, s/*/\*/
07:59damas has joined #asahi-dev
08:00unifiedgravity[m]Noice looks like we finally have a proxy uart
08:02chaos_princessjannau: btw, i think the firmware hacks will have to stay, i already tried feeding 14.8.3 OTA to installer, and it also fell over on wifi fw collection
08:04unifiedgravity[m]Okay I have the proxy connected to my M2 Max so I think we'll call it a day and pick up tomorrow. Thanks for all your patience and support.
08:08damas has quit [Ping timeout: 480 seconds]
08:26chadmed has quit [Quit: Konversation terminated!]
09:09kaazoo has joined #asahi-dev
09:34kaazoo has quit [Ping timeout: 480 seconds]
09:34Leo3418 has quit [Quit: Applying updates]
09:44kaazoo has joined #asahi-dev
09:54pb17 has quit [Ping timeout: 480 seconds]
09:55Leo3418 has joined #asahi-dev
09:56pb17 has joined #asahi-dev
10:10jannaufl0_id: does m1n1's cpufreq init work for you? for me applying cpu-fixed-freq-pll-relock on perf cores breaks/hangs
10:11fl0_idjannau no. In all my tests I disabled cpufreq, because I could never figure out why it hangs
10:20dataviruset has quit [Quit: dataviruset]
10:25maettu102 has quit [Read error: Connection reset by peer]
10:25maettu102 has joined #asahi-dev
10:51kaazoo has quit [Ping timeout: 480 seconds]
11:08pb17 has quit [Ping timeout: 480 seconds]
11:11kaazoo has joined #asahi-dev
11:21pb17 has joined #asahi-dev
12:16chadmed has joined #asahi-dev
12:58c10l has joined #asahi-dev
13:03c10l has quit [Quit: Bye o/]
13:03c10l has joined #asahi-dev
13:30nst has quit [Ping timeout: 480 seconds]
13:41nst has joined #asahi-dev
13:51voidah has quit [Remote host closed the connection]
13:51voidah has joined #asahi-dev
14:49picnoir has quit [Quit: WeeChat 4.9.0]
15:03arryn has joined #asahi-dev
15:05arrynwhen installing a raw kernel boot object. is there more to be done than just using kmutil to install it? on M5 pro, I can't even get a payload to run that just does b . it just boot loops
15:06chaos_princesswhich macos version?
15:06chaos_princessin that case just bputil + kmutil should be enough
15:07chaos_princess -nckas
15:17arrynthe machine I have is a T6050 and I installed a 2nd macOS partition to keep my original one intact
15:18jannauwith an endless loop as start the second mac isn't really helpful. I was going to ask for serial logs with EARLY uart
15:23arrynmhhh for some reason I don't hear the boot chime. is it a global setting or per partition?
15:23chaos_princessnvram, so global
15:25arryneven though I specified the offset for the start function. ill try to put jumps into my exception vectors
15:26voidah has quit [Remote host closed the connection]
15:26voidah has joined #asahi-dev
15:29junefish has quit [Remote host closed the connection]
15:29junefish has joined #asahi-dev
15:30chaos_princessno, it jumps to 2048
15:30chaos_princessthis aligns so the next is at 2048 in file
15:33arrynmy kmutil command is the one from the wiki: kmutil configure-boot -c myobj.o --raw --entry-point 2048 --lowest-virtual-address 0 -v /Volumes/macOS2/
15:34chaos_princessok, but is myobj.o what you think it is
15:34chaos_princesshexdump it
15:35arrynwill do. it should be, I used the same toolchain and program I developed on M1, just put a loop at the start to see if I can get it to rum
15:41kaazoo has quit [Ping timeout: 480 seconds]
15:43arrynfor objdump of try the m1n1 elf I get LLVM ERROR: Invalid data was encountered while parsing the file. bur hexdump of m1n1 shows the vector table at 0 and start at 0x800
15:44arrynis it time to conclude that apple broke something again? if the chime finishes and the macOS kernel boots, that means the firmware on my partition is fine too, right?
16:24chadmed has quit [Ping timeout: 480 seconds]
16:24chadmed has joined #asahi-dev
16:37voidah has quit [Remote host closed the connection]
16:38voidah has joined #asahi-dev
16:56damas has joined #asahi-dev
17:00dcavalca856979766785894 has quit [Read error: Connection reset by peer]
17:00dcavalca856979766785894 has joined #asahi-dev
17:04damas has quit [Ping timeout: 480 seconds]
17:21DarkShadow4444 has joined #asahi-dev
17:28DarkShadow44 has quit [Ping timeout: 480 seconds]
17:31jsk has quit [Remote host closed the connection]
17:38jsk has joined #asahi-dev
17:44damas has joined #asahi-dev
17:52damas has quit [Ping timeout: 480 seconds]
18:02apc8 has joined #asahi-dev
18:02apc is now known as Guest8723
18:02Guest8723 has quit [Quit: Ping timeout (120 seconds)]
18:02ohtwoone has quit [Quit: Ping timeout (120 seconds)]
18:02apc8 is now known as apc
18:03ohtwoone has joined #asahi-dev
18:17NSA_RECON_BOT has quit [Ping timeout: 480 seconds]
18:25NSA_RECON_BOT has joined #asahi-dev
18:36NSA_RECON_BOT has quit [Ping timeout: 480 seconds]
18:44NSA_RECON_BOT has joined #asahi-dev
18:45ieyasu has quit [Remote host closed the connection]
18:55ieyasu has joined #asahi-dev
18:59NSA_RECON_BOT has quit [Ping timeout: 480 seconds]
19:07NSA_RECON_BOT has joined #asahi-dev
19:24ieyasu has quit [Remote host closed the connection]
19:24ohtwoone has quit [Quit: Cya]
19:48opticron has quit [Read error: Connection reset by peer]
19:48ohtwoone has joined #asahi-dev
19:49opticron has joined #asahi-dev
19:57voidah has quit [Ping timeout: 480 seconds]
19:58voidah has joined #asahi-dev
20:13c10l has quit [Ping timeout: 480 seconds]
20:46floklisven: how do I know which macOS version is running / was used to install into the second partition? mac mini m4 pro here, I get the boot chime, a progress bar, and then it drops me into a slideshow with support.apple.com/macsetup at the bottom, instead of running m1n1
20:47chaos_princessboot into other macos, do "get info" in finder at the volume
20:48floklithe 'choose startup disk' window apparently also shows the version. both are listed as macOS 26.4.1
20:49flokliI ran `kmutil configure-boot -c m1n1-bla.bin --raw --entry-point 2048 --lowest-virtual-address 0 -v /Volumes/m1n1`
20:51flokliThen `csrutil disable`, cause it complained. ignored the 'need to reboot for changes to take effect'. ran again the kmutil command. /macsetup error screen, rebooted into 1tr again, and tried to run kmutil again. still same behaviour
20:52chaos_princessbputil -nckas
20:53kaazoo has joined #asahi-dev
21:00pb17 has quit [Ping timeout: 480 seconds]
21:03first_thunder has quit [Quit: The Lounge - https://thelounge.chat]
21:03pb17 has joined #asahi-dev
21:04first_thunder has joined #asahi-dev
21:15jannauI pushed initial M3 Pro and M3 Max/Ultra device trees to asahi-wip. doesn't do much besides booting to serial console
22:12hardy94 has joined #asahi-dev
22:32damas has joined #asahi-dev
22:40damas has quit [Ping timeout: 480 seconds]
23:05pb17 has quit [Ping timeout: 480 seconds]
23:06pb17 has joined #asahi-dev
23:14hardy94 has quit [Ping timeout: 480 seconds]
23:36voidah has quit [Ping timeout: 480 seconds]