#
…

2022-07-11

00:20nsklaus_ has joined #asahi-dev
00:23nsklaus has quit [Ping timeout: 480 seconds]
00:28millenialhacker has joined #asahi-dev
00:29phire has quit [Remote host closed the connection]
00:31phire has joined #asahi-dev
00:36millenialhacker has quit [Ping timeout: 480 seconds]
01:00derzahl has joined #asahi-dev
01:15chengsun has joined #asahi-dev
01:20chengsun_ has quit [Ping timeout: 480 seconds]
01:28AProgrammer has joined #asahi-dev
01:29AProgrammerWas able to install Asahi although its not immediately obvious to me how to build the kernel as I see no .config in the /boot directory?
01:30AProgrammerThat is - build the kernel targeting AARCH_APPLE or whatever the correct platform name for M1 is - I'm not quite sure
01:31tpw_rules/proc/config.gz is in it
01:32ditlood has joined #asahi-dev
01:36AProgrammer has quit [Remote host closed the connection]
01:40chengsun_ has joined #asahi-dev
01:46chengsun has quit [Ping timeout: 480 seconds]
01:51millenialhacker has joined #asahi-dev
01:58chengsun has joined #asahi-dev
01:58nuh^ has quit [Ping timeout: 480 seconds]
01:59millenialhacker has quit [Ping timeout: 480 seconds]
02:01chengsun_ has quit [Ping timeout: 480 seconds]
02:03chengsun_ has joined #asahi-dev
02:06nuh^ has joined #asahi-dev
02:07chengsun has quit [Ping timeout: 480 seconds]
02:10chengsun has joined #asahi-dev
02:13chengsun_ has quit [Ping timeout: 480 seconds]
02:14chengsun has quit [Read error: Network is unreachable]
02:14chengsun has joined #asahi-dev
02:27chengsun has quit [Quit: Quit]
02:27chengsun has joined #asahi-dev
02:38nuh^ has quit [Remote host closed the connection]
02:51chengsun_ has joined #asahi-dev
02:53chengsun has quit [Ping timeout: 480 seconds]
02:54saina has joined #asahi-dev
02:55saina has quit []
02:57dsrt^ has joined #asahi-dev
03:00phiologe has joined #asahi-dev
03:04PhilippvK has quit [Ping timeout: 480 seconds]
03:05chengsun_ has quit [Ping timeout: 480 seconds]
03:06chengsun has joined #asahi-dev
03:29nicolas17 has quit [Quit: Konversation terminated!]
03:32millenialhacker has joined #asahi-dev
03:36chengsun has quit [Ping timeout: 480 seconds]
03:36chengsun has joined #asahi-dev
03:40dsrt^ has quit [Ping timeout: 480 seconds]
03:41millenialhacker has quit [Ping timeout: 480 seconds]
03:46chengsun_ has joined #asahi-dev
03:47dsrt^ has joined #asahi-dev
03:48chengsun has quit [Ping timeout: 480 seconds]
03:54chengsun_ has quit [Read error: Connection reset by peer]
03:54chengsun has joined #asahi-dev
04:21chengsun_ has joined #asahi-dev
04:26chengsun has quit [Ping timeout: 480 seconds]
05:13dsrt^ has quit [Ping timeout: 480 seconds]
05:13mps has quit [Quit: leaving]
05:15mps has joined #asahi-dev
05:28millenialhacker has joined #asahi-dev
05:35doggkruse has quit [Ping timeout: 480 seconds]
05:37millenialhacker has quit [Ping timeout: 480 seconds]
05:53chadmedhmph neofetch is going to *suck* to make work properly with big.LITTLE systems
05:55chadmedits just an enormous shell script so it will require some pretty annoying changes to how it enumerates the cpus
06:03tpw_rulesi thought as long as the picture shows up that's all that matters :)
06:04chadmedheh true but "(10) @ 2.064 GHz" looks stupid and is wrong to boot
06:04chadmedbut its going to be too painful to fix so im not gonna bother
06:05chadmedkf5solid and lscpu were easy
06:06chadmedthe style rules for neofetch make it almost impossible to efficiently support proper identification of aarch64 parts anyway
06:06chadmedyoure not allowed to pipe anything and youre meant to avoid external utilities altogether...
06:06chadmedso it would be an absolute mess of case/esac and evil awk regex matching
06:37millenialhacker has joined #asahi-dev
06:40JTLIf neofetch is the largest thing you need to worry about I'd consider that a small problem :P
06:40chadmedwell yeah i dont really care at this point, if the patches i sent to util-linux and kde get merged then thats good enough for me
06:41chadmedi kind of want to hack minimal DT support into kinfocenter's main view but i feel like it would be messy and not well received
06:41chadmedcurrently it only supports getting information from DMI
06:49millenialhacker has quit [Ping timeout: 480 seconds]
06:50the_lanetly_052 has joined #asahi-dev
06:55gladiac is now known as Guest4717
06:55Guest4717 has quit [Read error: Connection reset by peer]
06:55gladiac has joined #asahi-dev
06:57gladiac has quit [Read error: No route to host]
06:57gladiac has joined #asahi-dev
07:09mps has quit [Remote host closed the connection]
07:09mps has joined #asahi-dev
07:50millenialhacker has joined #asahi-dev
07:58mps_ has joined #asahi-dev
07:58millenialhacker has quit [Ping timeout: 480 seconds]
08:00mps has quit [Ping timeout: 480 seconds]
08:08bisko has joined #asahi-dev
08:25the_lanetly_052 has quit [Remote host closed the connection]
08:26millenialhacker has joined #asahi-dev
08:34millenialhacker has quit [Ping timeout: 480 seconds]
08:40the_lanetly_052 has joined #asahi-dev
08:41bisko has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
09:04the_lanetly_052 has quit [Remote host closed the connection]
09:04the_lanetly_052 has joined #asahi-dev
09:16MajorBiscuit has joined #asahi-dev
09:48millenialhacker has joined #asahi-dev
09:56millenialhacker has quit [Ping timeout: 480 seconds]
10:06marcanasahi branch pushed, now based on 5.19-rc5 and with preliminary t8112 support
10:07marcanseems to work on my t6000, hopefully there are no regressions. the reboot notifier stuff got merged with significant changes, so that series is now dropped and macsmc is ported over.
10:07marcanpovik: minor fixup to macaudio too
10:08marcanerr... USB broke?
10:08marcanso much for no regressions
10:09j`ey$ git rev-list 88084a3df167..asahi/asahi | wc -l
10:10j`eyhow does USB break?
10:10marcanjust plays dead
10:11marcanI probably broke old DART
10:15bisko has joined #asahi-dev
10:22marcanwait now I'm confused, it worked
10:24marcanit works now... I'm lost
10:36jannauit works with the usb device plugged already before boot
10:39marcanit's always been plugged in
10:40povikmarcan: did you pick up upstream admac?
10:40poviknot sure if it's in 5.19-rc5
10:40* povikis checking
10:40marcandon't think so
10:41povikah not yet there
10:41povikthat's good actually
10:41povikthe binding changed
10:43jannauusb-reset was one thing which could have explained working/broken usb in some situations
10:46povik(it's overdue that we update the binding, but as i understand it marcan you have a flight to catch :p)
10:46marcanit was erroring out on controller init, but it seems just that one time?
10:47marcanyeah er, I'm going to try to get a -dev everything release out today and if nothing is broken promote to prod after the flight
10:47marcanwell, if nothing regresses anyway
10:47sveni've seen that when messing with the usb3 mess but I hope you haven't merged that yet :D
10:48marcanbluetooth neither, but I can add that if you want
10:48svenideally someone would test bluetooth on a m1 max/pro/ultra first
10:48svenbut other than that it should be fine
10:49marcanI'll do that, I always test on m1pro by default (it's my build machine)
10:50jannauI thought it was known broken with the newer wifi/bt modules
10:51sven_older_ actually :D
10:51svenit crashes the T2 wifi/bt module after exchanging a few HCI commands
10:52svenM1 pro/max/ultra should work since that's what R did their reverse engineering with
10:53marcanokay, let's go for it. where's the branch?
10:54marcan(just pushed -asahi1 packages to asahi-dev if anyone wants to play along with the testing, but let's add BT and call it -asahi2)
10:54marcanalso I need to clean up the t8112 devtrees a bit... and add the air speculatively :)
10:54svenhttps://github.com/AsahiLinux/linux/tree/bluetooth-wip, you'll need to add the board-type for the t6xxx ones though since I based it on the latest RC
10:55svenand you probably also want my latest m1n1 PR which byte-swaps the BD address because of course it has to be the other byte order than mac addresses
10:55svenanything else would be too easy, can't have that *sigh*
10:57marcanI remember that from the wiimote days...
10:57wuhoipok[m] has joined #asahi-dev
11:05marcanpunting on the keyboard for now, we need to investigate that
11:06jannau5.19.0-rc5-asahi-1-1-ARCH boots on the mac studio and usb works
11:06marcan(hint: if anyone wants to throw the MTP firmware into ghidra and see if there's any sane way to shut it down such that it spits out the HID descriptors again on the next startup, that would be most helpful)
11:06svenoh, right, the wiimotes were Bluetooth as well
11:10marcanjannau: what keyboard layout do you have, btw?
11:12jannaugerman, was the only option for release day delivery
11:14kov has joined #asahi-dev
11:15marcanjannau: can you pull m1n1 and run experiments/mtp.py and paste the output? (you need to chainload)
11:15marcanjust want to see if there is something that looks like keyboard countrycode with a different mapping
11:16marcan(bbiab, quick dinner)
11:26marcan[MTP][stm] 00000000 14 01
11:26marcan[MTP][stm] 00000000 14 02
11:26marcan... who wants to bet 00 = ANSI, 01 = ISO, 02 = JIS
11:26marcanI suspected something like that
11:27marcanI might just go with that, hope I guessed right. saves me from having a nasty translation table in m1n1...
11:29bisko has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
11:57kettenisI should put the m2 patches on the u-boot asahi branch now that I reviewed them ;)
12:01bisko has joined #asahi-dev
12:07joske has joined #asahi-dev
12:09joskemarcan: I'm getting: drivers/iommu/apple-dart.c:1268:79: error: macro "DART_TCR" requires 2 arguments, but only 1 given
12:10marcanoh, I think that's with PM_SLEEP
12:10marcanI forgot I missed that
12:13dsrt^ has joined #asahi-dev
12:13kettenismarcan: pushed the m2 patches onto the u-boot asahi branch in case that helps
12:13marcanit does! thanks!
12:13kettenisthe branch is up-to-date with u-boot master, which means it is pretty much u-boot 2022.07 which is supposed to be released today
12:14kettenisnot on that branch are:
12:14kettenis1. the nvme shutdown patch (you really need to send that one upstream)
12:14kettenis2. jannau's patch to pick the paired ESP
12:15j`eymarcan: will we need APPLE_RTKIT_HELPER for m1 machines?
12:18bisko has quit [Ping timeout: 480 seconds]
12:20_jannau_j`ey: maybe, not for the currently supported HW
12:20j`eyok thanks
12:22marcandon't think so
12:32bisko has joined #asahi-dev
12:34MajorBiscuit has quit [Quit: WeeChat 3.5]
12:36kettenisthat reminds me; I did look at the mtp/dockchannel diff over the weekend
12:37kettenisso APPLE_RTKIT_HELPER is just a driver that brings up RTKit and processes log messages on the "standard" endpoints?
12:38MajorBiscuit has joined #asahi-dev
12:39dsrt^ has quit [Ping timeout: 480 seconds]
12:40bisko has quit [Ping timeout: 480 seconds]
12:41the_lanetly_052__ has joined #asahi-dev
12:46bisko has joined #asahi-dev
12:47the_lanetly_052 has quit [Ping timeout: 480 seconds]
12:52marcanI suspect I will need some MTP<->RTKit cooperation at some point for suspend/resume
12:52marcanbut for now this works
12:53marcanthe problem is handoff to linux
12:53marcanfor keyboard only dockchannel is trivial, you can pretty much hardcode an init message and start getting HID reports
12:53marcanthough ideally you'd process the init messages and look for the right interface
12:55nsklaus has joined #asahi-dev
12:55nsklaus__ has joined #asahi-dev
12:56nsklaus_ has quit [Read error: Connection reset by peer]
12:57dsrt^ has joined #asahi-dev
12:58kettenissince the boot picker uses the keyboard/touchpad, there must be a way to reinitialize
12:59_jannau_the bootpicker reboots
12:59chadmedhow does that work for booting into paired recovery tough
13:00chadmedit doesnt reboot if you do that
13:00Gaspare has joined #asahi-dev
13:00chadmedor is the boot picker an app inside the paired recovery
13:00nsklaus has quit [Read error: Connection reset by peer]
13:01_jannau_yes, the boot picker is the initial app of the paired recovery
13:11joske has quit [Remote host closed the connection]
13:12joske has joined #asahi-dev
13:12millenialhacker has joined #asahi-dev
13:15Gaspare has quit [Quit: Gaspare]
13:16joske has quit [Remote host closed the connection]
13:17joske has joined #asahi-dev
13:19Gaspare has joined #asahi-dev
13:20millenialhacker has quit [Ping timeout: 480 seconds]
13:26Gaspare has quit [Quit: Gaspare]
13:31marcankettenis: do not underestimate just how shitty apple firmware designs can be :p
13:31marcanthis isn't new
13:31j`eyreseting seems to have been a problem since day 1
13:37Gaspare has joined #asahi-dev
13:55kettenisu-boot 2022.07 was released so I rebased the u-boot asahi branch once more to pick up the Makefile changes that drops -rc6
14:00joske has quit [Remote host closed the connection]
14:00Gaspare has quit [Quit: Gaspare]
14:15dsrt^ has quit [Ping timeout: 480 seconds]
14:16dsrt^ has joined #asahi-dev
14:17Gaspare has joined #asahi-dev
14:20Catyre has joined #asahi-dev
14:35millenialhacker has joined #asahi-dev
14:36balrog has joined #asahi-dev
14:37Gaspare has quit [Quit: Gaspare]
14:42dsrt^ has quit [Ping timeout: 480 seconds]
14:43millenialhacker has quit [Ping timeout: 480 seconds]
14:44dsrt^ has joined #asahi-dev
14:47dsrt^ has quit [Remote host closed the connection]
14:49bisko has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
15:03Catyre has quit [Remote host closed the connection]
15:07balrog has quit [Ping timeout: 480 seconds]
15:09Gaspare has joined #asahi-dev
15:31Catyre has joined #asahi-dev
15:33balrog has joined #asahi-dev
15:46millenialhacker has joined #asahi-dev
15:50mps_ is now known as mps
15:50saina has joined #asahi-dev
15:54millenialhacker has quit [Ping timeout: 480 seconds]
15:55saina has quit []
16:02doggkruse has joined #asahi-dev
16:02Gaspare has quit [Quit: Gaspare]
16:03c10l has quit [Quit: Bye o/]
16:07mps_ has joined #asahi-dev
16:09mps has quit [Ping timeout: 480 seconds]
16:36marcansven: I see the newer BT firmwares end in 'es2'
16:36marcanFILE brcm/brcmbt4387c2-apple,madagascares2-u.bin SHA256 d18a3cf25d8a24638d81ff81e60b5ae48c79ec1f5602ee3fab7764d8756b5c1e
16:36marcanFILE brcm/brcmbt4387c2-apple,madagascares2-u.ptb SHA256 754c5e3d73c9c9e07a5e54bb8386f01269c2c1016cf5237ed1ebe66a697c9971
16:36marcanFILE brcm/brcmbt4387c2-apple,maldiveses2-u.bin SHA256 a6bc9668240087301f55b52377f906f296e0fcfff841dd3d4bc08121e6ad7ea6
16:36marcanFILE brcm/brcmbt4387c2-apple,maldiveses2-u.ptb SHA256 992bb635f267c385a4ead8d217a34256a4e59dc2d6c7586f627b3c99b373333b
16:36marcanFILE brcm/brcmbt4387c2-apple,okinawaes2-u.bin SHA256 1dccd2749e374f791aa73457b6d9458a60bf9c3eeb3d1f322bd9d5fe9c51a54a
16:36marcanFILE brcm/brcmbt4387c2-apple,okinawaes2-u.ptb SHA256 77853c87c0bab51be2039694b850526644a3e450eee2d64cdf69989e9adc375d
16:36marcanshould we put that into the board name for bluetooth, or drop it in the firmware extractor?
16:39svenhuh, I assume the WiFi board type is without that es2 then?
16:40svenmacos itself just has a static list of filenames for bt fwiw
16:40marcanlovely...
16:40marcanI say drop it, I think it's BS and not part of the real platform name
16:40Erik52 has joined #asahi-dev
16:40svenI think I’d drop it…
16:40Erik52 has quit []
16:40svenyeah, that
16:40marcanbesides, we don't control the ACPI name and it wouldn't have it
16:41svendunno if you saw, but the Bluetooth function has no acpi companion on the T2 macs
16:42svenso it’s either hardcode “formosa” since there’s only a single FW anyway or do some hacks to get to the WiFi acpi name
16:43marcanah, we can just hardcode it then, heh
16:43svenyeah, I don’t expect them to release another t2 Mac ;)
16:49jole has quit [Remote host closed the connection]
16:49jole has joined #asahi-dev
17:08Catyre has quit [Remote host closed the connection]
17:08MajorBiscuit has quit [Ping timeout: 480 seconds]
17:10marcansven: pacman upgrade on asahi-dev for some magic :)
17:11marcan(building kernel with bluetooth right now, that's not there yet)
17:15Gaspare has joined #asahi-dev
17:27nicolas17 has joined #asahi-dev
17:32marcan[ 7.237106] hci_bcm43xx 0000:01:00.1: enabling device (0000 -> 0002)
17:32marcan[ 7.237734] hci_bcm43xx 0000:01:00.1: Reading OTP failed with -2
17:32marcan[ 7.237735] hci_bcm43xx: probe of 0000:01:00.1 failed with error -2
17:33svencan you enable the dev_dbg prints?
17:33svenI guess the otp might’ve moved on the max/pro then
17:34marcanthe wlan one did
17:35svenor maybe just dump the entire BAR0 from the HV, that's how I found it in the first place :D
17:35j`eymarcan: did you intentionally remove rtkit-helper in asahi-5.19-rc5-2?
17:36marcandid -1 have it?
17:36marcanI didn't touch the config
17:36j`eyno, I mean the file disappeared
17:37Catyre has joined #asahi-dev
17:37marcanthe module?
17:37marcanoh like in the merge?
17:37j`eyfatal: Path 'drivers/soc/apple/rtkit-helper.c' exists on disk, but not in 'asahi-5.19-rc5-2'.
17:37marcanohh wait I know what that is
17:37marcanI dropped the dockchannel branch because that stuff ended up in the hidspi one anyway
17:37marcanforgot rtkit helper was only there
17:37marcanI'll put it back
17:38marcanthanks :)
17:38j`eynp, I always 'git diff asahi-5.19-rc5-1 asahi-5.19-rc5-2' when i see a new branch ;)
17:42dingodoppelt has quit [Quit: ZNC 1.9.x-git-170-9be0cae1 - https://znc.in]
17:42dingodoppelt has joined #asahi-dev
17:42marcan00000005c14140e0 84cc0000 00000038 00000000 00000000 00000000 00000000 00000000 00000000
17:42marcan00000005c1414100 00000000 00000000 00000000 00000000 00000000 00000000 00000000 80004433
17:42marcan00000005c1414120 00000002 18001ff9 00000000 80005f71 00000002 19328ff5 00000000 00081d15
17:42marcan00000005c1414140 3d730000 4d003243 4850543d 3d6d204e 20372e34 3d562020 80ff0075 060c8309
17:42marcan00000005c1414160 70955119 0000ae00 00000000 00000000 00000000 00000000 00000000 00000000
17:42marcan00000005c1414180 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
17:42marcanthat's regmon swapped, but you get the idea
17:45marcan413c maybe?
17:45Catyre has quit [Ping timeout: 480 seconds]
17:48bisko has joined #asahi-dev
17:51svenyeah, could be
17:52sven(Not at a computer here, staring at those numbers on my phone :D)
17:56marcan[bluetooth]# list
17:56marcanController F8:4D:89:61:CE:97 macherie [default]
18:05svenguess that just leaves those weird T2 chips
18:14the_lanetly_052__ has quit [Ping timeout: 480 seconds]
18:18millenialhacker has joined #asahi-dev
18:20anoake[m]This will help jail break the T2 chips
18:20marcanthat's of no use here
18:20marcanwe're not trying to jailbreak anything
18:21marcanT2 Macs are already open machines as far as the AP is concerned
18:21marcanwe're talking about the bluetooth chipset
18:23svenpushed the OTP offset thingy
18:23sven(plus some minor cleanup)
18:26millenialhacker has quit [Ping timeout: 480 seconds]
18:29marcansven: I already did that lol
18:29marcan-3 is already pushed
18:30sventhat also works
18:30sventhat driver needs a bit more work before I'll submit it anyway
18:30doggkruse has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
18:31sveni'd mainly like to figure out suspend and how to reset this thing
18:32marcanyeah well wifi is kind of broken there too..
18:32marcanfrom what I saw, it's Yet Another Bespoke Thing for wifi
18:32marcansome other command mechanism or something, instead of the little mailbox thing brcm normally uses
18:33sveni only briefly looked at BT but it didn't seem to bad
18:33marcanwifi is probably just bad because that driver is a hairball to begin with
18:33marcanat least with bt it's your driver :>
18:33svenfair :D
18:33svenI think suspend is mostly just tear down the communication rings + write 0 to a register that R called "sleep control"
18:34doggkruse has joined #asahi-dev
18:35j`eyweird, 'git fetch --all' doesn't give me asahi-5.19-rc5-3
18:35sveni'm probably also missing some kind of "ack this irq" poke because if you switch to legacy interrupts it'll trigger an irq storm
18:35marcanj`ey: it's totally there...
18:35j`eymarcan: I know, I see it in the github UI!
18:37marcansven: okay, I tried pairing my phone and it exploded :p
18:37marcanI see a lot of [ 96.167889] Bluetooth: hci0: Unknown advertising packet type: 0x2713
18:37svenhuh, how?
18:37marcan[ 56.933345] Bluetooth: hci0: sending frame failed (-22)
18:37marcanpairing failed
18:37marcanand now the list is empty
18:38svenive seen the first thing as well, that should be harmless
18:38marcanconnecting fails
18:38sventhere’s a decent chance it died
18:38marcanit does let me try again if I remove the device
18:38doggkruse has quit []
18:39marcanand goes through the pin dance so it's not dead
18:39marcanbut pairing fails
18:39svenhuh, that's new
18:39svenusually when you do something wrong it just dies
18:39sven-22 is EINVAL, let's see
18:39marcansec, this *could* be my fault
18:41marcanJul 12 03:40:51 macherie bluedevil-wizard[986]: kf.bluezqt: PendingCall Error: "br-connection-profile-unavailable"
18:41marcanmaybe this is just bluetooth stupidity
18:41svenlet me add some more warnings around return -EINVAL anyway
18:41marcanI do get that error when pairing fails though
18:41marcanso it seems related
18:42marcanfailed with my speakers too
18:44svenlet me try to re-pair something
18:45marcancould be a 4387 thing
18:46svenlooks like it, pairing with my phone still works here
18:49marcan[ 43.841986] too big 1032 0 1028
18:50marcansorry, had my own crap debug :D
18:50marcanit's the length
18:50svenhuh, which ring is that?
18:50svenmaybe I messed up the payload size somewhere
18:50marcanlet me use your patch
18:51svenlooks like ACL
18:51svenI guess just bump it by another 4 bytes and call it a day, lol? :D
18:52marcanI don't get how it can be that big but ok lol?
18:52svenme neither :D
18:52svenjust s/HCI_MAX_FRAME_SIZE/HCI_MAX_FRAME_SIZE + 4/ and i guess it'll work
18:54marcanjust for completeness
18:54marcan[ 185.661718] hci_bcm43xx 0000:01:00.1: payload len 1032 is too large for ring 5 (max is 0 or 1028)
18:54svenyeah, that’s acl host to device
18:56svenguess this is the time where I finally have to take a look at the Bluetooth spec
18:59AProgrammer has joined #asahi-dev
18:59AProgrammerHow can I install the linux kernel headers in asahi?
19:01marcansven: still fails
19:01marcanmight be unrelated to the tx failure tbh
19:01marcanthis could just be userspace fail
19:02svenhrm, this is pairing with an Android phone?
19:02marcanyes, but also happened with some speakers, let me try them again
19:02Catyre has joined #asahi-dev
19:02svenpairing so far has as always worked here fwiw, usually the userspace fail happened afterwards
19:03marcanyeah I think it's just the UI being stupid
19:03marcanit does actually pair possibly
19:04marcanoh yeah, it works
19:04marcanjust pulseaudio was down so it refused to connect lol
19:05svenah yes, that sounds like userspace fail I ran into as well :D
19:05marcanyeah, obex works and stuff
19:06svennice! I could never get obex to work because apparently macOS doesn’t really support that out of the box
19:06nicolas17I once actually thought Linux didn't support my laptop's bluetooth chipset when in fact I hadn't installed pulseaudio-module-bluetooth x_x
19:06marcanah wait, probably just because I startx'ed instead of starting a real session
19:06marcan(this is my weirdo test machine)
19:07marcanok, let me push another kernel with the +4 anyway, and see if it magically works on my cleaner machines
19:07svenI did play a sine wave from macOS and sent that to the m1 though via that audio ACL thing :D
19:07svenstill couldn’t really recognize it in btmon though :(
19:08nicolas17sven: I think "Applications/Utilities/Bluetooth File Exchange.app" does OBEX
19:08svenoh, I didn’t find that one
19:09nicolas17it's kinda hidden nowadays, "cool kids use AirDrop"
19:09svenjust some 10+ year old Python based thing that wouldn’t run anymore and some crappy App Store apps that didnt do what they claimed to do
19:10nicolas17yeahhh I don't think iOS ever gave third-party apps access to Bluetooth at the level needed to implement OBEX
19:11doggkruse has joined #asahi-dev
19:11nicolas17there's many apps like that which actually only work with another instance of itself on the other phone
19:12nicolas17... oh you mean mac app store? that's more plausible
19:15marcansven: did you see my firmware shenaigans btw? :>
19:15marcanjust pacman -Syu should make it work
19:15marcan*shenanigans
19:17svenI, uh, don’t actually have the asahi distro installed and kinda still boot from sown random rootfs :D
19:18svenmostly because I’ve been too lazy to hook up screen/keyboard/mouse and do the while 1TR dance again
19:18svengot a link though? I wanna see what you did :>
19:20AProgrammer has quit [Remote host closed the connection]
19:20svenah, found it
19:20svenasahi-installer
19:21marcan(and the PKGBUILDs to go with it)
19:21svenit contains a asahi_firmware Python package now and I…
19:21svenyeah… and I assume there’s a marching PKGBUILD ;)
19:21marcanasahi-fwextract
19:21sven*matching
19:21marcanwhich has a postinstall that runs it
19:22svenvery nice
19:22marcanand all of which is now a dep of asahi-base
19:22marcan*asahi-meta
19:22marcan... and the bluetooth stuff a dep of asahi-desktop-meta
19:22marcanincluding a postupgrade that enables bluetooth.service
19:22marcanI want the bluetooth icon to just pop up for people :D
19:23svennice, so everything just magically works :>
19:23j`eysven: almost makes you want to install the asahi distro :P
19:23povikit does :p
19:24svenI’ll switch my main machine over once thunderbolt works
19:24marcanalso the installer should do M2 now (requires expert mode, the 12.4 for that stub is one of the magic special M2 versions, will not meet prereqs on "normal" 12.4 machines)
19:25marcanbut I haven't pushed/tested it yet
19:25Catyre has quit [Remote host closed the connection]
19:25sventhat’s a hard dependency in my setup and the only reason I even decided to get entangled with that mesa
19:25j`eyand studio too
19:25Catyre has joined #asahi-dev
19:26j`eysven: looking forward to testing BT, will be nice to get rid of the headphone cable :D
19:26svenit was fun to write a nice self contained driver again. pretty much the opposite of the thunderbolt mess :)
19:26j`eymarcan: I presume youre not going to push a new installer until you're back?
19:27marcanyou presume wrong
19:27marcanI want to get the dev one pushed today
19:27blazra[m]it does just work! just updated and rebooted and it works (14" m1pro with BT headphones), thanks!
19:27j`eythere was someone asking about m1 ultra a few days ago
19:27marcanand if people say it works I can promote to prod during the trip no problem
19:27j`eymarcan: cool
19:27j`eyblazra[m]: fast feedback!
19:27marcandidn't even give *me* time to test it on a proper install :D
19:28marcanalso the kernel is still missing the +4
19:28marcanasahi4 should be up now
19:30jannaumarcan: uboot/m1n1 (and kernel) from asahi-dev work on the m2 (not a proper install but asahi on an usb stick). kernel not fully tested since I can't convince mkinitcpio to include all necessary modules
19:31doggkruse has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
19:31marcanyes, I expected that
19:31marcanwas just going to look now
19:32marcanoh btw, for the installer, not sure if you all saw ./test.sh
19:32Catyre has quit [Remote host closed the connection]
19:32Catyre has joined #asahi-dev
19:32marcanassuming you have a proper dev env on macos (notably, you need the rust stuff for m1n1 to build, but you could cheat by copying over a built tree), you can just test the installer in-place
19:32jannauhaven't looked yet at the installer
19:33svenBluetooth on M2 would be interesting as well. If it has a new chip revision the driver maybe needs a chance for sending the calibration data
19:33jannaulol, in my cursed usb install we're not using it
19:33marcanI think it's the same rev?
19:33marcanjannau: using what?
19:34svenoh, maybe j was confused. I thought M2 was 4387 c3 while the pro/max are 4387 c2
19:35svenand possibly c3 also needs the beam forming calibration. that thing again is a hard coded list somewhere in macOS :/
19:35marcanah, wait, M2 *pro* is 4378 r3
19:35marcanM2 *air* I think is just the same 4387
19:36svenyeah, than maybe it needs an additional check in that send_calibration as well. Or maybe not
19:36jannaumarcan: the asahi initcpio hook but nervermind, seems to be just my cursed setup
19:44marcanjannau: pushed an asahi-scripts bump for that
19:50doggkruse has joined #asahi-dev
20:02marcantrying to make new OS images now, let's see if they work on M2
20:05jannauit should, packages from asahi-dev work on m2
20:09marcanyup, bluetooth Just Worked on my (no nonsense end user install) iMac
20:14marcanOS images built, going to take a shower while they upload
20:26Catyre has quit [Remote host closed the connection]
20:27Catyre has joined #asahi-dev
20:33Catyre has quit [Remote host closed the connection]
20:33Catyre has joined #asahi-dev
20:33saina has joined #asahi-dev
20:33saina has quit [Read error: Connection reset by peer]
20:48Catyre has quit [Remote host closed the connection]
20:49Catyre has joined #asahi-dev
20:53marcaninstalling on the M2...
20:53jannaumarcan: mkinitcpio hooks needs "map add_firmware apple/tpmtfw-j413.bin apple/tpmtfw-j493.bin"
20:54marcanno it doesn't
20:54marcandockchannel-hid.c needs MODULE_FIRMWARE ;)
20:55jluthra has quit [Remote host closed the connection]
20:55jluthra has joined #asahi-dev
20:57Catyre has quit [Ping timeout: 480 seconds]
21:00leitao has joined #asahi-dev
21:01marcanbut really this is a problem for redistribution
21:01marcanso I need to find a solution for that
21:02mps has joined #asahi-dev
21:03marcanI think the driver should probably have some kind of retry logic, so that it picks up the firmware once booted
21:03marcansince the keyboard will work perfectly fine without it
21:04mps_ has quit [Ping timeout: 480 seconds]
21:05leitao has quit [Remote host closed the connection]
21:10marcanwhoops, u-boot failed to boot RTKit NVMe. that's annoying.
21:10marcan"apple_rtkit_boot returned -110"
21:10marcanisn't that a timeout?
21:11marcandidn't we increase those?
21:11jannauyes, I did increase the timeouts
21:11marcanoh well, I tried
21:11marcanI need sleep
21:15marcanfeel free to try to debug it if you have time...
21:15marcanI may need to punt on this for a few days, I'm not sure I'll have time before the trip
21:16jannausleep well, m2 nvme works for me in u-boot (chainloading m1n1+u-boot over usb)
21:16marcanrelease u-boot does not
21:17marcanconfigs/etc are in the PKGBUILDs repo
21:19marcanis it supposed to go 1...0... boot, or is it supposed to boot immediately at 0?
21:19marcanI see 1 second from 0 to the error, either that's the 1 second timeout, or it isn't actually waiting 1 second
21:19jannauI'm chainloading the boot.bin from the system
21:20jannaubut m1n1 doesn't touch the nvme in that setup
21:20marcantry actually chainloading
21:21marcanmake -C .. CHAINLOADING=1 && python tools/chainload.py -r ../build/m1n1.bin && cat ../build/m1n1.bin <(echo -ne 'chosen.asahi,efi-system-partition=a17b7e46-e950-bb4f-bc82-8ab1047a058e\nchainload=a17b7e46-e950-bb4f-bc82-8ab1047a058e;m1n1/boot.bin\n') >| /tmp/m1n1-chainload.bin && python tools/chainload.py -r /tmp/m1n1-chainload.bin
21:21marcan(replace with your ID)
21:21marcananyway, good night :)
21:22marcanit's possible I broke it with the m1n1 rtkit power stuff in stage1, and it's broken for new installs on m1 too
21:22marcanbut I've been chainloading like that and it worked for me, I think?
21:22marcanso I'm not sure
21:27jannauwith m1n1 chinloading from the nvme it's broken as well
21:34jannauit's not broken on m1 ultra
21:42Catyre has joined #asahi-dev
21:50Catyre has quit [Ping timeout: 480 seconds]
21:55jannaufixed by resetting nvme via pmgr on probe
21:59Catyre has joined #asahi-dev
22:14Catyre has quit [Remote host closed the connection]
22:15Catyre has joined #asahi-dev
22:19joske has joined #asahi-dev
22:19joskepovik: marcan: new asahi branch, no audio, I got this in dmesg: firmware designs can be :p
22:19joske<marcan> this isn't new
22:19joske<j`ey> reseting seems to have been a problem since day 1
22:19joske<kettenis> u-boot 2022.07 was released so I rebased the u-boot asahi branch once more to pick up the Makefile changes that drops -rc6
22:20jannaudoesn't cause problems on the m1 ultra
22:21joskethat was some palm x-select, I meant: [ 18.077498] apple-mca 238400000.mca: ASoC: error at snd_soc_dai_set_sysclk on mca-i2s-2: -16
22:22joske has quit [Remote host closed the connection]
22:23Catyre has quit [Ping timeout: 480 seconds]
22:34joske has joined #asahi-dev
22:34joskehmm if I boot previous kernel same problem, even when reverting boot.bin, so please disregard for now
22:35joske has quit [Remote host closed the connection]
22:38Gaspare has quit [Quit: Gaspare]
22:43Catyre has joined #asahi-dev
22:55c10l has joined #asahi-dev
22:58Catyre has quit [Remote host closed the connection]
22:58Catyre has joined #asahi-dev
23:06Catyre has quit [Ping timeout: 480 seconds]
23:11Catyre has joined #asahi-dev
23:15psykose has joined #asahi-dev
23:19millenialhacker has joined #asahi-dev
23:27millenialhacker has quit [Ping timeout: 480 seconds]
23:30Catyre has quit [Remote host closed the connection]
23:31Catyre has joined #asahi-dev
23:39Catyre has quit [Ping timeout: 480 seconds]
…