#
…

2026-04-26

00:09freetolga has joined #asahi
00:39vimproved has quit [Remote host closed the connection]
00:41vimproved has joined #asahi
00:42marceljj has quit [Ping timeout: 480 seconds]
01:00chadmed has quit [Quit: Konversation terminated!]
01:00chadmed has joined #asahi
02:14freetolga has quit [Quit: Connection closed for inactivity]
02:16hexdump01 has joined #asahi
02:18hexdump0815 has quit [Ping timeout: 480 seconds]
02:26tobhe_ has joined #asahi
02:33tobhe has quit [Ping timeout: 480 seconds]
02:35Volvagia356 has joined #asahi
02:41anticataclyst has joined #asahi
02:44Brainium has quit []
03:25wetdog has joined #asahi
03:28wetdog has quit []
03:41anticataclyst has quit [Ping timeout: 480 seconds]
03:47marvin24 has joined #asahi
03:59nightstrike has quit [Quit: Connection closed for inactivity]
05:18Game1145401954 has quit []
05:19Game1145401954 has joined #asahi
06:18___nick___ has joined #asahi
06:18___nick___ has quit []
06:20___nick___ has joined #asahi
07:02Volvagia356 has quit [Ping timeout: 480 seconds]
07:32pb17 has quit [Ping timeout: 480 seconds]
07:52pb17 has joined #asahi
08:17Volvagia356 has joined #asahi
08:22thelounge42 has quit [Quit: The Lounge - https://thelounge.chat]
08:23thelounge42 has joined #asahi
08:28HerbY_NL has joined #asahi
08:43alarumbe has quit []
08:57Yushi has quit [Remote host closed the connection]
08:58Yushi has joined #asahi
09:08riker77 has quit [Quit: Quitting IRC - gone for good...]
09:20Axenntio has joined #asahi
09:20Axenntio has quit []
09:23riker77 has joined #asahi
09:55thelounge42 has quit [Quit: The Lounge - https://thelounge.chat]
10:00monthebiff has joined #asahi
10:00monthebiffooooo a blog post
10:12monthebiffoooooo power stuff
10:13leioSomething is seriously off in the power section of the report
10:13leionamely, PMGR is NOT a three letter acronym
10:22thelounge42 has joined #asahi
10:23monthebiffso power stuff and vrr... yay!
10:24jbemaybe there's a ligature in that three letter acronym
10:24jbeor it is an extended three letter acronym
10:25HerbY_NL has quit [Quit: Leaving]
10:25thelounge42 has quit []
10:26thelounge42 has joined #asahi
10:29travltux[mds]chadmed (IRC): for gentoo, I can only find the gnome-extra/iio-sensor-proxy package. via https://bugs.gentoo.org/565904 Oliver Freyermuth indicates that:
10:29travltux[mds]"This is also an optional runtime dependency of dev-qt/qtsensors and required for Ambient Light Sensor support for automatic screen brightness control starting with Plasma 6.6.0."
10:29travltux[mds]"So probably the category name should be adapted, as it is not Gnome-specific?"
10:29travltux[mds]is that the correct package to install?
10:30chadmedyeah, im trying to avoid having to package it in ::asahi. i just built it from source and installed it to test stuff
10:30chadmedit's currently not in ::gentoo and given that it's load-bearing for plasma and gnome i dont think we're the right people to maintain it
10:30chaos_princessit hopefully should be packaged into ::gentoo eventually, but i don't think maintaining it is exactly in asahi's scope
10:31chaos_princessthough i may just go ahead and do it
10:31travltux[mds]chadmed (IRC): okydoke, thx for letting me know and your wonderul post <3
10:31ravikant_ has joined #asahi
10:31travltux[mds]chaos_princess (IRC): <3
10:31chaos_princess(as soon as my j314 becomes usable, it has been running elixir's indexer for 3 days already and i am afraid to touch it)
10:33chaos_princessthat is with free-threaded python, and after i've replaced berkeleyDB with rocksdb (it made the process go faster)
10:33chadmedyou run elixir locally?
10:33chaos_princesshosted version does not have our branches
10:34travltux[mds]TIL about elixir
10:34chaos_princessalso, there indexer leaks and is now at 25gb used
10:38chadmedat least it doesnt seem too resource hungry once the indexing is done
10:38chadmedannoying that limiting the set of tags requires modifying code though
10:51ravikant_ has quit [Ping timeout: 480 seconds]
11:11breakgimme has quit [Remote host closed the connection]
11:11breakgimme has joined #asahi
11:18Brainium has joined #asahi
11:19Moprius has joined #asahi
11:47artun[m] has joined #asahi
11:50artun[m]Hey there, I'm trying to install asahi with new 0.8.0 installer. never had this issue before. tried the disk utility gui method, rebooted, still can't install.
11:52chaos_princessreboot into recovery, run disk utility and scan all devices and partitions using that disk utility app.
12:02chaos_princessdo the thing it tells you to
12:11Game1145401954 has quit []
12:12Game1145401954 has joined #asahi
12:12artun[m]i’m not sure what flags to put into fsck (and what file is container instead of /dev/rdisk3s1).
12:12artun[m](running first aid in disk utility for the partition is not giving me an error, only this partition)
12:13Game1145401954 has quit []
12:15Game1145401954 has joined #asahi
12:15chaos_princessin gui disk utility, click on "view", then on "show all devices"
12:15chaos_princessyou will be able to run it on a full container that way
12:16___nick___ has quit []
12:18___nick___ has joined #asahi
12:18___nick___ has quit []
12:20___nick___ has joined #asahi
12:42artun[m]it worked. thank you =)
12:57monthebiffthe PMP stuff in the blog, is there a list of validated machines somewhere... and a quick how to on updating device trees floating around?
12:58chaos_princessm1 pro/max, i think base m2 too
13:02EliSaado[m]1the how to is in the blog right?
13:02EliSaado[m]1it mentions rebooting into macOS and running the new installer to update the firmware
13:02monthebiffi keep thinking about the power stuff... that side of stuff is so cool to me. Nerdy cool. But cool.
13:03EliSaado[m]1hehe I agree
13:03EliSaado[m]1and a 20% improvement is genuinely huge
13:04monthebifftheres an example in the blog for updating firmware - but in the energy section it explicitly mentions updating a device tree to enable the power stuff as its shipping not enabled by default.
13:05monthebiff20% is massive
13:05EliSaado[m]1oh you're right
13:05EliSaado[m]1I mangled the firmware and device tree sections in my head
13:06EliSaado[m]1best bet might be to check the device tree definitions on the branch that chaos was working on if that's public at all
13:07monthebiffreading the asahi docs it says the ADT is in binary format... so there must be a tool for interacting with it. (The fact that im asking this question probably means i should just wait until it ships upstream)
13:08chaos_princesshttps://github.com/AsahiLinux/linux/blob/asahi-wip/arch/arm64/boot/dts/apple/t6001-j314c.dts - just stick a `#define APPLE_USE_PMP 1` here (well, the one that corresponds to your device), then rebuild kernel
13:09monthebiffthere any power savings tied to the VRR stuff as well? kinda just hoping there is
13:09EliSaado[m]1oh chaos already linked it
13:13fl0_idEliSaado[m]1 but you need to compile to binary after editing
13:15EliSaado[m]1I'm on a patched kernel anyway (for fairydust support) so i have to compile the kernel about once a week :S
13:15monthebiffright ok, so involves compiling a new kernel? like the full linux kernel? feel like im mis understanding
13:15j`eyjust the device tree, not the full kernel
13:17fl0_idthough if you haven't compiled it before and do not explicitly specify only dtbs, it will compile everything...
13:17monthebiffand in this context... 'compiling' the device tree means converting to binary? Not like compiling a program into an executable and linking etc etc?
13:17j`eyyeah converting to binary
13:17chaos_princessj`ey: i think it is only in -wip rn, so full kernel
13:17monthebiffhuh. cool.
13:18monthebiffone day im gonna have to start compiling stuff. my webdeveloper shoes are showing.
13:18j`eyoh ok, needs the driver compiled too, when I read about the ifded in the dts i assumed the pmp driver was enabled
13:20monthebiffjust in case it got lost in the noise, any power savings stuff with the VRR stuff? (Fingers crossed thinking lower refresh rates more of the time more power savings etc.)
13:23bennofs has quit [Read error: No route to host]
13:26bennofs has joined #asahi
13:27mohit815822635306907 has quit [Remote host closed the connection]
13:28mohit815822635306907 has joined #asahi
13:41MichaelLongthx for the new progress report, it was a good read as always!
13:54___nick___ has quit [Remote host closed the connection]
14:00Game1145401954 has quit []
14:05Moprius has quit []
14:09Game1145401954 has joined #asahi
14:11Yushi has quit [Remote host closed the connection]
14:12Yushi has joined #asahi
14:24chadmed has quit [Read error: Connection reset by peer]
14:24chadmed has joined #asahi
15:01BlueMountainPilgrim[m] has joined #asahi
15:29EtherNet has joined #asahi
15:29xx has joined #asahi
15:31turboretor has joined #asahi
15:33xx has quit [Quit: Konversation terminated!]
15:39monthebiffis there a way to access the proton11-arm beta in steam atm?
16:01Guest8052 has joined #asahi
16:08user_ has joined #asahi
16:09jotenakis2 has joined #asahi
16:11user_hi, im trying to figure out why my m1 almost always fails to boot on kernel 6.18-6.19. i compared a successful boot on 6.19 to 6.17 and these lines only appear on 6.19: https://paste.sh/8l9hExdg#-lD5I5D07UixdMWhIMd0r3yz . are these normal for newer kernels or is it the actual cause?
16:13user_also, not sure if this helps, but the only way i managed to get it to somewhat reliably boot was by plugging in a charger on 6.17, updating and rebooting
16:14jannauall three look unrelated. the first two is missing calibration data for the ambient light sensor
16:15Guest8052 has quit [Ping timeout: 480 seconds]
16:16jannauthe third one is returning EPROBE_DEFER. either just a probe order issue or missing driver. Even in the missing driver case this should not be fatal
16:17jannauuser_: which device? how dos the the failure to boot look like? black screen (with a white dash in the left top corner, possibly blinking? Is this a Fedora-Asahi-Remix system?
16:17user_its the same symptoms/hardware as issue #465, but its hard to figure out what causes it because the whole device just freezes at a random point in boot
16:18chaos_princessjust as a test, can you create a second install and see if that boots?
16:18user_im using void linux but the person who made the issue was using fedora. it also freezes before luks unlock so i dont think that can be related
16:20user_im not sure if i can make a second install, my macos is shrunk as far as it goes
16:21chaos_princess`export EXPERT=1` before running
16:21chaos_princessjust remember to delete and expand macos afterwards
16:22user_i kind of already did that for my first install
16:22user_could the logs in /boot/asahi help?
16:23chaos_princessidk? i want to check something specific and making a clean install is the easiest way to do that
16:24user_the install wasnt that long ago, was around 12/2025
16:24chaos_princessthe thing i am interested in happened between 6.17 and 6.18
16:26user_what was the change? i could try manually extracting it and adding it to the current install
16:27chaos_princessi remember something changing with the dts in 6.18, and remember bug reports where dts blobs were failing to be updated
16:28chaos_princessso if you can mount the linux partition and ensure m1n1.bin has the latest dts, that would be great
16:28user_/boot/dtbs has the 6.19 one, it was extracted by the package manager
16:28user_how do i check whether it was added to m1n1.bin
16:29chaos_princessidk, i guess binwalk it in extract mode and see what it finds there
16:30zaxl has quit [Ping timeout: 480 seconds]
16:32user_the post install hook has this, cat ${DTBS} >>"${TARGET}.new", and DTBS is set to the right path in the config
16:33chaos_princessright, but did it actually do it
16:33chaos_princessthat is the question
16:36anticataclyst has joined #asahi
16:36user_i added extra logging to the hook this was the command being run: cat /boot/dtbs/dtbs-6.19.12+1-asahi_2/apple/*.dtb >> "/boot/m1n1/boot.bin.new"
16:37chaos_princessdid it replace boot.bin with the file that generated?
16:37chaos_princesswhat is the mod time on that?
16:38user_well if it didnt it wouldnt be booting 6.19
16:38user_im using the one where it directly boots kernel
16:38user_instead of uboot
16:38chaos_princesscan you try it with a real u-boot
16:39chaos_princessit probably won't change anything, but just to confirm that it is not due to your setup being weird
16:42user_ has quit [Quit: leaving]
16:48user_ has joined #asahi
16:49user_not sure if i did something wrong but i couldnt get out of uboot
16:49user_boot failed and nvme scan showed nothing
16:49chaos_princesswell, thats unexpected
16:50chaos_princessasahi u-boot fork, right?
16:50user_yeah AsahiLinux/u-boot
16:51chaos_princessi am not sure what to do next tbh
16:52chaos_princessaside from the original proposal of a second install, at least to check if it is broken at the start, or the system ended up in a weird state that broke it
16:53user_there weren't any errors during install if thats what you mean
16:54chaos_princessno, i want to check with a system that i know to be good
16:55user_i know my setup is kind of weird but the fact that someone else with a standard fedora install had the same issue rules that out i feel
16:56chaos_princessthere may be issues during package updates or something similar
16:58user_its also not like the package is completely broken because it does sometimes successfully boot and there arent any issues after that
16:59chaos_princesssure, but the point is that i don't have any better ideas
17:00user_the person who opened this issue also had another issue i did where using the lower usbc port on <=6.17 causes the system to freeze
17:00user_though that was fixed in 6.18 when it successfully booted
17:02user_i feel like since symptoms are similar its related to rework of usb stuff
17:02user_but also ive never looked at code so that might be nonsense
17:09user_ has quit [Quit: leaving]
17:20ravikant_ has joined #asahi
17:39anticataclyst has quit [Ping timeout: 480 seconds]
17:52HerbY_NL has joined #asahi
17:54monthebiff has quit [Quit: leaving]
17:55c10l has quit [Quit: Bye o/]
17:57dataviruset has joined #asahi
17:57HerbY_NL has quit []
17:57anticataclyst has joined #asahi
17:58ravikant_ has quit [Ping timeout: 480 seconds]
18:02zaxl has joined #asahi
18:08ravikant_ has joined #asahi
18:09maettu102 has quit [Quit: brb]
18:10ravikant_ has quit []
18:12maettu102 has joined #asahi
18:14daniel_ has joined #asahi
18:15daniel_Hi, since VRR is mentioned in the latest blog post, I wanted to check on my MacBook Pro M1 to see how many more watts 120Hz draws compared to 60Hz. Interestingly, Btop (I don’t know if the wattage displayed there is reliable) shows the lowest power consumption at 120Hz (about 5.5 W) and the second-lowest at 50Hz (about 5.7 W). All other refresh rates are around 6.1 W. That’s strange, isn’t it?
18:15daniel_ has quit []
18:20pitust has quit [Read error: Connection reset by peer]
18:20pitust has joined #asahi
18:21jotenakis2 has quit [Quit: Konversation terminated!]
18:21jotenakis2 has joined #asahi
18:23dataviruset has quit [Quit: dataviruset]
18:23WindowPain has quit [Remote host closed the connection]
18:25WindowPain has joined #asahi
18:43gi0 has joined #asahi
18:46gi0 has quit []
18:56c10l has joined #asahi
18:58Hevea[m] has joined #asahi
19:22rvalue- has joined #asahi
19:27rvalue has quit [Ping timeout: 480 seconds]
19:48marvin24 has quit [Ping timeout: 480 seconds]
20:00Game1145401954 has quit []
20:00Game1145401954 has joined #asahi
20:09OMGasm has quit [Remote host closed the connection]
20:18marshmallow_ has quit [Ping timeout: 480 seconds]
20:22snow-racoon has joined #asahi
20:31snow-racoon has quit [Ping timeout: 480 seconds]
20:45freetolga has joined #asahi
21:16jotenakis2 has quit [Quit: Konversation terminated!]
21:17jotenakis2 has joined #asahi
21:21Volvagia356 has quit [Ping timeout: 480 seconds]
22:01kaazoo has joined #asahi
22:05kaazoo has quit []
22:18jotenakis2 has quit [Remote host closed the connection]
22:52thelounge42 has quit [Quit: The Lounge - https://thelounge.chat]
22:57thelounge42 has joined #asahi
23:10mountainlion has joined #asahi
23:14mountainlionI have a macbook neo, how can I help yall with A18 Pro development. can run tests and report back
23:16mountainlionhad a gross misunderstanding thinking apple silicon was fully supported by asahi. might get rid of it for an M1 or M2 macbook for now. but can follow instructions for now to help anyway i can, I ain't much of a dev
23:16chaos_princessneo has the same problem that need solving as m4, and someone else is already working on it using a m4. Not much to help with right now, sorry.
23:18mountainlionit may help them since i have access to the A18 version, it still lets me disable SIP. I can definitely do things outside the norm on macos w it
23:19linkmauve has quit [Remote host closed the connection]
23:19linkmauve has joined #asahi
23:20chaos_princesssip is irrelevant, everything can disable it, what matters is SPTM, and neither a18 nor m4 have pre-sptm versions of macos
23:20revadam has joined #asahi
23:20BrontyBuutveld[m] has joined #asahi
23:21revadam has quit []
23:22BrontyBuutveld[m]James Calligeros: got an incredible deal on an m2 macbook air, installing asahi now
23:24mountainlionoh heck. they've screwed reverse engineers with that. you can't even read those sections of memory from what I am reading, even as root.
23:26chaos_princessthere is a solution, it is just a lot of work. basically load our thing before sptm, and then run it in a separate vm next to macos. not easy but doable
23:26mountainlionbut on apple's documentation it says the M2-M4 all utilize that. So I assume SPTM was implemented by about macos 15
23:28mountainlionwell, i could do that to gather data if i was told how. but otherwise imma just get an M1 or M2 macbook. the aggressiveness of macos with enforcing daemons that I have no want or need for is something I hate as a user.
23:29chaos_princess15 for all things m4 and base m2, 26 for m3 and m2 pro/max/ultra, 26.4 for m1 pro/max/ultra
23:31mountainlionwas a massive pain in the ass to neuter siri/spotlight/intelligence, and even still I was only able to fully disable a few daemons.
23:32mountainlionlooked into repacking tahoe myself then i find an encrypted apfs image that i can't touch.
23:32nicolas17modifying the filesystem is doable but a pain in the ass
23:33mountainlionhad mostly stayed away from apple devices, but people i know in opsec/it strongly recommended them to me, then i find it's mostly an immutable black box and you are just trusting apple to not do anything nefarious in that blackbox
23:34chaos_princessthat describes pretty much anything with dma-capable firmware :P
23:35freetolga has quit [Quit: Connection closed for inactivity]
23:37mountainlionthe claim is for security. but when you blackbox half the system you may prevent third parties from finding CVEs, but that exact thing done by whitehats is a primary method CVEs are discovered. the lack of transparency is a concern imho
23:40JamesCalligeros[m]it's less of a concern when the system architecture is designed to disallow many of common attack vectors from the ground up. everything is behind its own IOMMU, SPTM/PPL does a decent job of preventing kernel pages from being pwned, etc.
23:40linkmauve has quit [Remote host closed the connection]
23:40linkmauve has joined #asahi
23:40JamesCalligeros[m]there isnt a single modern, performant desktop-class system available commercially that doesnt have some level of "trust me bro" required. as far as "trust me bro" goes, the security model of these devices is about as good as you can hope for too
23:41JamesCalligeros[m]if youre that concerned about low level black boxes, go buy a raptor power9 board
23:43mountainliona third party is very unlikely to get into your mac, but it is shocking to me the high level of telemetry that still goes on in the background even with privacy features maxxed out in the system settings app. LuLu would be popping up every 15min showing a random apple service I don't even use still trying to phone home.
23:43nightstrike has joined #asahi
23:45mountainlionand i agree with everything you just said. you don't 'own' any device unless you made the chips yourself lol
23:48JamesCalligeros[m]even with telemetry, youre going to struggle to find a vendor that doesnt do it. it's not possible. at the very least apple pay lip service to privacy and dont access to it, unlike some others
23:49JamesCalligeros[m]anything identifiable goes to icloud pcc apparently anyway. anything else is probably just metrics
23:51chaos_princesseven we do telemetry :P (opt-in, ofc)
23:51JamesCalligeros[m]s/dont access/dont sell access
23:52chaos_princessif you enable ADP, they can't access your icloud stuff at all
23:55nicolas17"the high level of telemetry that still goes on" compared to what? :P
23:56mountainlionit is nothing compared to an android device with google play that's for sure
23:57mountainlionthe privacy rabbit hole isn't worth my time that's what i'm learning. but started with a pihole and when you see a flood of google telemetry 24/7 it opens up a rabbit hole
23:58JamesCalligeros[m]theres a happy medium and it's somewhere between "let google know your bowel movements" and "go live in a cave with the morelocks"
23:58JamesCalligeros[m]i think as far as american tech vendors in 2026 go, apple's approach is probably the best you can hope for
23:58mountainlionthe only way to make a device secure is to airgap it, turn it off, remove any batteries, lock it in a safe, and cover it in concrete
…