#
…

2026-08-21

00:53pb17 has quit [Ping timeout: 480 seconds]
00:54pb17 has joined #asahi-dev
01:09___nick___ has quit []
01:11___nick___ has joined #asahi-dev
01:11___nick___ has quit []
01:13___nick___ has joined #asahi-dev
02:00yurix has quit [Remote host closed the connection]
02:00yurix has joined #asahi-dev
02:37tobhe_ has joined #asahi-dev
02:44tobhe has quit [Ping timeout: 480 seconds]
03:12bgb has quit [Remote host closed the connection]
03:12nora has joined #asahi-dev
03:14nora_ has quit [Ping timeout: 480 seconds]
03:16bgb has joined #asahi-dev
03:28bgb has quit [Remote host closed the connection]
03:37bgb has joined #asahi-dev
04:00pb17 has quit [Remote host closed the connection]
04:01pb17 has joined #asahi-dev
04:07Lyude has joined #asahi-dev
04:15Lyude has quit [Quit: Bouncer restarting]
04:18Lyude has joined #asahi-dev
04:26Lyude has quit [Quit: Bouncer restarting]
04:29Lyude has joined #asahi-dev
04:40lyudess has joined #asahi-dev
04:42Lyude has quit [Ping timeout: 480 seconds]
05:43aurora has quit [Remote host closed the connection]
05:43aurora has joined #asahi-dev
06:11mildsunrise has quit [Quit: disconnect]
06:11mildsunrise has joined #asahi-dev
06:57hardy94 has joined #asahi-dev
07:02rosalyntg has quit [Read error: Connection reset by peer]
07:02rosalyntg has joined #asahi-dev
07:09bennofs3 has quit [Ping timeout: 480 seconds]
07:10bennofs3 has joined #asahi-dev
07:32sveni wonder if we should just skip the pipehandler lock when switching to dummy on teardown
07:32svenbecause i always get the "lock not acked" there
07:37hardy94 has quit [Quit: hardy94]
07:55hardy94 has joined #asahi-dev
07:55hardy94 has quit []
07:56hardy94 has joined #asahi-dev
08:17bgb has quit [Remote host closed the connection]
08:17bgb has joined #asahi-dev
08:24aurora has quit [Remote host closed the connection]
08:25aurora has joined #asahi-dev
08:32chadmed has quit [Quit: Konversation terminated!]
10:33pb17 has quit [Ping timeout: 480 seconds]
10:42pb17 has joined #asahi-dev
10:48picnoir has quit [Ping timeout: 480 seconds]
10:57picnoir has joined #asahi-dev
11:08JamesCalligeros[m]jannau: re that top surface, it is fine with argb and abgr too when src width is the width of the screen
11:09JamesCalligeros[m]Height is once again immaterial
11:09JamesCalligeros[m]P010 is busted however
11:10JamesCalligeros[m]I think this surface is just unusable for ~cursors
11:11rttdev has joined #asahi-dev
11:13JamesCalligeros[m]Every supported format except nv12 and p010 is fine lolwut
11:17rttdev has quit [Quit: bye]
11:20bennofs3 has quit [Ping timeout: 480 seconds]
11:29bennofs3 has joined #asahi-dev
11:45JamesCalligeros[m]Aha, it’s the scaler
11:45JamesCalligeros[m]Or at least is related to it
11:50chaos_princesswe aren't even powering it up, do we?
12:00JamesCalligeros[m]dcp does its own scaling
12:00JamesCalligeros[m]but disabling scaling makes this less broken
13:08rany has quit [Remote host closed the connection]
13:08rany has joined #asahi-dev
13:18sventhat only leaves the rest of the owl: pcie and dp tunnels :D
13:18svenoh, and dt-bindings
13:19chadmed has joined #asahi-dev
13:22rany has quit [Remote host closed the connection]
13:22rany has joined #asahi-dev
13:30nela has joined #asahi-dev
13:48eConnah has joined #asahi-dev
13:49eConnah has quit []
13:52hardy94 has quit [Remote host closed the connection]
13:52svenalso stumbled across a register called "AUSBC_PWR_CTRL" which we just completely ignore so far which seems to maybe be used in suspend/resume of atcphy
14:33pb17 has quit [Ping timeout: 480 seconds]
14:37pb17 has joined #asahi-dev
14:38jannausven: does it depend on anything in linux-next? rebase on v7.2 ran into following oops on unplug: https://paste.centos.org/view/fb114397
14:39svenhuh, don't think so. i just based it on torvalds/master
14:39jannautbt dock with an usb3 realtek network chip
14:40svenlet me check
14:40jannauerr, s:linux-next:torvalds/master:
14:40sveni don't think it does, i just rebased on that because i wanted to get rid of the nvme commits
14:41sveni did some cleanup in thunderbolt/apple.c though, maybe that broke something
14:43jannauat least connected to the dock usb3 fallback of an asm2464 usb4 ssd enclosure worked
14:46hardy94 has joined #asahi-dev
14:47svenhm, i wonder if it's related to the usb3 quirk i had to add. don't see how though
14:50jannauI don't think so. when connecting it directly it shows now up as highspeed device so your uab2 fix works there
14:51svenoh, i mean your kernel panic
14:54tryhard[m]someone can help me figure out this crash? https://pastebin.com/QRtCh6k2
14:54tryhard[m]the crash is happening because pmp rtkit fails to wake up, not sure why? some pointers would help
14:55tryhard[m]didn't push some changes because they dont matter that much
14:55jannauthe kernel panic seems to require 3 levels of usb4 hubs/docks for reproduction
14:56sventhat sounds like pcie hotplug hell!
14:57jannauthis time it was a random pointer in klist_next+0x1c/0x1b8
14:58jannauhotunplug hell
14:58svenmaybe you're just hitting this now because the usb3 tunnels are actually up now
14:59jannauI did not try multiple docks before
14:59svencan you show me the other kernel panic?
14:59svenand maybe run with KASAN or however it's called
15:00jannauwithout usb3 tunnelling the netdev wouldn't exists
15:01jannauI'll try kasan
15:01ipatch has quit [Quit: WeeChat 4.9.5]
15:02svendrivers/thunderbolt RiiR wen? :-P
15:06sveni think i just reproduced it with just a single dock and tbnet behind it
15:07tryhard[m]ok i might've fixed that crash, but now this crash happens https://pastebin.com/NzJj4v97
15:08tryhard[m]some other driver tries to probe but fails
15:09jannaupredictable network names from hell: "r8152 4-1.1.2.4.1.4:1.0 enu1u1u2u4u1u4: renamed from eth0"
15:11jannaupanic doesn't reproduce with kasan
15:15jannauneeds tbnet, the macbook went to sleep and so it was missing
15:16chadmed has quit [Quit: Konversation terminated!]
15:16chadmed has joined #asahi-dev
15:16svenyeah, this smells like something with xdomain goes wrong, there's tb_xdomain_unregister in the stack tace
15:16ipatch has joined #asahi-dev
15:20svenlooks like something has been freed already by the time the xdomain is torn down
15:20chaos_princesshow exactly does one parse the "remap" for darts?
15:21sveni got it wrong from staring at the ADT, janne just dumped the REMAP register where it was more obvious
15:22sventhe switch is already gone but somehow the xdomain teardown still wants to do something with it
15:23jannaufrom dcp,acio,usb map ((remap >> 8) & 0xff) to (remap & 0xff)
15:24jannauthere was more complicated one
15:25jannauthe remap itself is simple and can be configured in probe. no good idea for dt-bindings though
15:27popey has quit [Ping timeout: 480 seconds]
15:28jannaut6020's dart-pmp has remap = 0xd0000000c, no idea how to interpret that
15:29svenyou have a x86 machine as well, don'y you? can you try to connect a dock and then another host and then unplug the dock?
15:31jannausure, I wouldn't bet on that it survices that
15:32svenyeah, same. looking at git log I suspect a8937f35cf39c39c64325aa84d0463d866850857
15:34svenif it happens on the x86 machine I’d probably just report it to the Thunderbolt maintainers
15:35sventhis sounds tricky to track down and fix
15:35chaos_princessthunderbolt0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>
15:36chaos_princessvery few phy changes needed
15:38chaos_princessand cio reset is even worse for statically figuring out the address
15:39jannauno panic on w86 with v7.1.8 but a8937f35cf39c39 is only in v7.2-rc1
15:52tryhard[m]is there any reason linux would want to power on dispext even when there is no external display?
15:59jannausven: no panic on 13th gen core
15:59jannaueven with v7.2
16:00svenoh, maybe it needs that driver unbind or root domain teardown the apple one does :/
16:03bennofs3 has quit [Quit: Ping timeout (120 seconds)]
16:03bennofs3 has joined #asahi-dev
16:06jannaudriver unbind runs into warnings in nhi_shutdown but no panic
16:07svenbah… so much for let someone else take care of that :D
16:14jannaumaybe it panicked on shutdown. shutdown resulted in frozen display and probably system
16:15jannauwhere is the uart on this laptop?
16:18JPeisachso i'm trying to locally merge in that tbt-reset-wip branch on top of asahi-wip.. that's a lot of merge conflicts.. am i doing something wrong
16:19svenyeah, that’s not gonna be fun
16:19svenit needs some tbt patches that are in 7.2
16:21JPeisachthanks. can't wait to start uni in two weeks and then have people say "git is hard" :)
16:22svengit _is_ hard and the UX is quite and imho. I’ve just gotten used to the pain
16:22sven*quite bad
16:22chaos_princessi may or may not have recorded a "how do i git" lecture at one of my previous places
16:23svenlinux with its dozens (hundreds?) of subsystem trees also results in extra fun
16:23svenespecially for hardware bringup that touches multiple subsystems
16:24JPeisachyeah thats the fun they don't know about.. anyways i'll come back when i got everything rebased
16:25chaos_princessi suggest doing the following - start with asahi-wip, merge usb-next, then either cherry-pick or interactive rebase tbt-reset-wip on top of the result
16:25JPeisachyep thats my plan
16:27svenyou do have thunderbolt devices or another thunderbolt-capable machine, right?
16:27JPeisachyep, i am back home and i got my monitor with me now
16:28fl0_idJPeisach maybe you can gift them the julia evans zine on git ;)
16:28svenyou're also aware the monitor won't work because that branch doesn't implement dp (or pcie) tunneling, right?
16:28chaos_princessunless it has a hub, it is the part of tb that is probably not happening soon
16:28JPeisachactually i believe i have a hub
16:29JPeisachjust need to borrow onne, lemme check
16:29svenonly thing that works is XDomain (machine to machine) and usb3-via-usb4 tunneling
16:30sventhe monitor will probably show up in boltctl and that's it if everything works
16:34JPeisachthis thing should work, right?
16:34chaos_princessnah, that looks usb3
16:35chaos_princessone way to figure out to support tb is to look at the price - if it is very expensive, it probably does tb
16:35jannauthe monitor probably has usb3 ports which might start to work
16:35JPeisachyeah it does
16:54rany has quit [Remote host closed the connection]
16:55rany has joined #asahi-dev
17:14JPeisachi think i'm about to bail, getting no success with the builds and errors around places - so i am afraid i am facing a skill issue
17:53balrog has quit []
17:55chaos_princesscan someone with m3 and/or m3pro check if tunable_AUX_SHM is a thing on your devices?
17:56balrog has joined #asahi-dev
18:07rzhikharevich has joined #asahi-dev
18:38svenjannau: yeah, pretty sure that commit is the issue
18:39sventb_switch_removes runs during unplug and freed the switches which are parents of the xdomain netdev and that xdomain thing only pins its direct parent
18:39svenso the grand parent is freed and then later that deferred works runs and calls that pm_runtime_whatever thing on the grandparent that's arleady free
18:39sven... i just have no idea how to fix that
18:46sofuschaos_princess: if that is checked by looking at the dt, then no for t6030
18:47roxfan has joined #asahi-dev
18:48chaos_princessneed live dt, not template
18:48chaos_princessshould be under the atc-phy node
18:50svendid you need any changes to the tbt driver for m3 btw? or did everything just work
18:50sofus/dev/mtd/by-name/adt should be good enough right? Thats what i used
18:51roxfan2 has quit [Ping timeout: 480 seconds]
18:52chaos_princessonly dt/phy/m1n1
18:52chaos_princesswell, i am having an issue with port 2 now, but both ports 1 and 0 work
18:52chaos_princesstbd if it is a dt issue or sth else
18:53chaos_princesssofus: yes, that should work
18:55chaos_princessdoes port 2 work on t602x? i have a suspicion it won't
18:56svenwhat's so special about port2? :D
19:01chaos_princessIncorrect mbox address in dts, it has F01100000 while F01108000 is correct
19:12sofusjust normal usb-c?
19:16svenif it works as-is on M3 i'll just add the compatibles to the initial submission as well
19:16beastd has quit [Ping timeout: 480 seconds]
19:17chaos_princessonly tested on the 'max. vanilla now also uses the pro/max style reset if i read it right
19:17svendid any m3 chip already have usb2v4? i suspect that will required changes but it may still be compatible enough
19:17chaos_princessand that has a new phy
19:18svenmeh, the reset binding is already reviewed and i just sent v2
19:18sveni'll leave that as-is but may as well already add the compatibles for acio/nhi then
19:19sveni'll probably get a m4 pro anyway, and m3 ultra is probably rare enough to just not bother
19:20chaos_princessm3 ultra likely has m4pro/max phy
19:20svenhow do you know it's new?
19:21chaos_princesskext is new. and yuka posted m4pro tunable list
19:22svenugh, so that's gonna be fun then
19:27jannauchaos_princess: t8122 has "tunable_AUX_SHM = None", t6030 doesn't have it at all
19:28chaos_princessso, thats even easier, it does not exist
20:06___nick___ has quit [Remote host closed the connection]
20:19yuyichao has quit [Remote host closed the connection]
20:29chaos_princessall 2 users of apple a11 now have to do an emergency update
20:37svenhahahaha, amazing :D
20:46chadmed has quit [Read error: Connection reset by peer]
20:46chadmed has joined #asahi-dev
21:14revadam has joined #asahi-dev
21:14revadam has quit []
21:16kettenisso a CVE just means a bug that should be backported now ;)
21:16kettenisbugfix I mean
21:19pb17 has quit [Ping timeout: 480 seconds]
21:36pb17 has joined #asahi-dev
22:37junefish has quit [Remote host closed the connection]
22:46vimproved has quit [Ping timeout: 480 seconds]
22:51remexre has quit [Ping timeout: 480 seconds]
23:28junefish has joined #asahi-dev
23:30remexre has joined #asahi-dev
…