#
…

2026-08-28

00:03robpatterson123 has quit [Ping timeout: 480 seconds]
00:04ddxtanx has quit [Remote host closed the connection]
00:21robpatterson123 has joined #asahi-dev
00:29robpatterson123 has quit [Ping timeout: 480 seconds]
00:32robpatterson123 has joined #asahi-dev
00:33pb17 has quit [Ping timeout: 480 seconds]
00:35robpatterson123 has quit [Read error: Connection reset by peer]
00:36robpatterson123 has joined #asahi-dev
00:40ddxtanx has joined #asahi-dev
00:40robpatterson123 has quit [Read error: Connection reset by peer]
00:41robpatterson123 has joined #asahi-dev
00:45robpatterson123 has quit [Read error: Connection reset by peer]
00:45robpatterson123 has joined #asahi-dev
00:49robpatterson123 has quit [Read error: Connection reset by peer]
00:50robpatterson123 has joined #asahi-dev
00:53robpatterson123 has quit [Read error: Connection reset by peer]
00:54robpatterson123 has joined #asahi-dev
00:57pb17 has joined #asahi-dev
01:01robpatterson123 has quit [Read error: Connection reset by peer]
01:09___nick___ has joined #asahi-dev
01:18rzhikharevich has quit [Ping timeout: 480 seconds]
01:20rzhikharevich has joined #asahi-dev
01:31nnym_ has joined #asahi-dev
01:32rzhikharevich has quit [Ping timeout: 480 seconds]
01:36nnym has quit [Ping timeout: 480 seconds]
01:40rzhikharevich has joined #asahi-dev
02:00yurix has quit [Remote host closed the connection]
02:00yurix has joined #asahi-dev
02:23jn has joined #asahi-dev
02:27jnn has quit [Ping timeout: 480 seconds]
02:29tobhe has joined #asahi-dev
02:36tobhe_ has quit [Ping timeout: 480 seconds]
02:50pb17 has quit [Ping timeout: 480 seconds]
02:57pb17 has joined #asahi-dev
03:16voidah has quit [Ping timeout: 480 seconds]
03:25voidah has joined #asahi-dev
03:33voidah has quit [Ping timeout: 480 seconds]
03:40hardy94 has joined #asahi-dev
03:41voidah has joined #asahi-dev
03:50voidah has quit [Ping timeout: 480 seconds]
03:58hardy94 has quit [Remote host closed the connection]
03:58voidah has joined #asahi-dev
04:02nora_ has joined #asahi-dev
04:04nora has quit [Ping timeout: 480 seconds]
04:06voidah has quit [Ping timeout: 480 seconds]
04:14voidah has joined #asahi-dev
04:22voidah has quit [Ping timeout: 480 seconds]
04:29ddxtanx has quit [Remote host closed the connection]
04:30hardy94 has joined #asahi-dev
04:31ddxtanx has joined #asahi-dev
04:31voidah has joined #asahi-dev
04:36Yushi_ has joined #asahi-dev
04:40voidah has quit [Ping timeout: 480 seconds]
04:41Yushi has quit [Ping timeout: 480 seconds]
04:43voidah has joined #asahi-dev
04:44Yushi_ is now known as Yushi
04:45hardy94 has quit [Remote host closed the connection]
04:48hardy94 has joined #asahi-dev
04:56voidah has quit [Ping timeout: 480 seconds]
05:04voidah has joined #asahi-dev
05:13voidah has quit [Ping timeout: 480 seconds]
05:21voidah has joined #asahi-dev
05:29voidah has quit [Ping timeout: 480 seconds]
05:37voidah has joined #asahi-dev
05:53voidah has quit [Ping timeout: 480 seconds]
06:01voidah has joined #asahi-dev
06:09voidah has quit [Ping timeout: 480 seconds]
06:17voidah has joined #asahi-dev
06:25voidah has quit [Ping timeout: 480 seconds]
06:34voidah has joined #asahi-dev
06:42voidah has quit [Ping timeout: 480 seconds]
06:45hardy94 has quit [Remote host closed the connection]
06:45voidah has joined #asahi-dev
06:54voidah has quit [Ping timeout: 480 seconds]
07:04voidah has joined #asahi-dev
07:15voidah has quit [Ping timeout: 480 seconds]
07:20voidah has joined #asahi-dev
07:28voidah has quit [Ping timeout: 480 seconds]
07:34cy8aer has quit [Read error: Connection reset by peer]
07:36voidah has joined #asahi-dev
07:44voidah has quit [Ping timeout: 480 seconds]
07:46voidah has joined #asahi-dev
07:47cy8aer has joined #asahi-dev
07:56voidah has quit [Ping timeout: 480 seconds]
08:03voidah has joined #asahi-dev
08:11bennofs3 has quit [Ping timeout: 480 seconds]
08:11voidah has quit [Ping timeout: 480 seconds]
08:18bennofs3 has joined #asahi-dev
08:20voidah has joined #asahi-dev
08:28voidah has quit [Ping timeout: 480 seconds]
08:36voidah has joined #asahi-dev
08:44voidah has quit [Ping timeout: 480 seconds]
08:47cy8aer has quit [Ping timeout: 480 seconds]
08:47voidah has joined #asahi-dev
08:56voidah has quit [Ping timeout: 480 seconds]
09:03cy8aer has joined #asahi-dev
09:03sofusi cant find any h264 videos with artifacts after my latest commit
09:03sofusfinding broken videos is a bit annoying tho :(
09:06voidah has joined #asahi-dev
09:07yukadid you check h264 10bit?
09:12sofusthanks, not i did not, did you have issues with 10bit?
09:14voidah has quit [Ping timeout: 480 seconds]
09:23voidah has joined #asahi-dev
09:23yukaWhen I tried playing a random file from my collection a few weeks ago, the output was completely garbled. Then I pasted some logs here, and someone said it was because the video was 10bit.
09:23yukaI can try it again now
09:24yukashould I apply your open PR before testing?
09:31voidah has quit [Ping timeout: 480 seconds]
09:35sofusyes, thanks
09:35sofusunless its 4:2:2 10bit, then it would still be broken
09:35cy8aer has quit [Remote host closed the connection]
09:37chadmed has quit [Remote host closed the connection]
09:37chadmed has joined #asahi-dev
09:39yukaHow can I check that?
09:40yukampv says: Format: yuv420p, Colormatrix: bt.709
09:40voidah has joined #asahi-dev
09:41sofusthen it would say Format: yuv422p10le or something, so should it should work
09:42yukaCould it also be a problem with libva-v4l2_request?
09:43sofusit could yes, is it still broken?
09:44yukaHave not built the kernel with your PR yet
09:44yukabut without your PR, it's broken
09:46yuka"modern art"
09:46cy8aer has joined #asahi-dev
09:48voidah has quit [Ping timeout: 480 seconds]
09:48voidah has joined #asahi-dev
09:51hardy94 has joined #asahi-dev
09:53noopwafelreal cga vibes :)
09:57voidah has quit [Ping timeout: 480 seconds]
10:00yukasofus: can you check your dms?
10:07voidah has joined #asahi-dev
10:15voidah has quit [Ping timeout: 480 seconds]
10:23voidah has joined #asahi-dev
10:26chadmed has quit [Read error: Connection reset by peer]
10:27chadmed has joined #asahi-dev
10:32voidah has quit [Ping timeout: 480 seconds]
10:40voidah has joined #asahi-dev
10:41yukaI applied your PR, and still the same issue. I have the same issue of the colors being completely wrong with youtube videos (vp9 via yt-dlp), so actually I think this is an issue somewhere in the chain (mpv <-> libva-v4l2_request).
10:49voidah has quit [Ping timeout: 480 seconds]
10:49voidah has joined #asahi-dev
10:50fl0_id_yuka it's a space odyssey alright :)
10:53yukait goes away with --vo=dmabuf-wayland
10:53yukavlc and firefox don't seem to have any problem (but not 1000% certain they are actually using hardware decoding)
10:58voidah has quit [Ping timeout: 480 seconds]
10:58yukaAlso goes away with --gpu-context=wayland
11:07voidah has joined #asahi-dev
11:15voidah has quit [Ping timeout: 480 seconds]
11:24voidah has joined #asahi-dev
11:30chadmed has quit [Quit: Konversation terminated!]
11:30chadmed has joined #asahi-dev
11:32voidah has quit [Ping timeout: 480 seconds]
11:41voidah has joined #asahi-dev
11:45nela has quit [Ping timeout: 480 seconds]
11:49voidah has quit [Ping timeout: 480 seconds]
11:50voidah has joined #asahi-dev
11:54nela has joined #asahi-dev
11:59voidah has quit [Ping timeout: 480 seconds]
12:07voidah has joined #asahi-dev
12:15hardy94 has quit [Quit: hardy94]
12:15voidah has quit [Ping timeout: 480 seconds]
12:23voidah has joined #asahi-dev
12:25sofusi have the same issue with the colors being wrong when using gpu or gpu-next, also with hwdec=v4l2request
12:27chadmedthe "colours being wrong" appears to be pixel format related. the chroma is offset from the luma rather than any one colour being "wrong"
12:29garrettgr has quit [Quit: The Lounge - https://thelounge.chat]
12:30garrettgr has joined #asahi-dev
12:31voidah has quit [Ping timeout: 480 seconds]
12:34chadmedthe problem becomes more obvious with low entropy images ^
12:37sofuscould it be the nv12 output is given to the gpu instead of gpu format
12:39chadmedi dont think the gpu's yuv handling has been thoroughly tested thus far
12:39chadmedbut that said, the dmabuf-wayland vo also exports an nv12 buffer
12:40voidah has joined #asahi-dev
12:41sofusyeah i dont really now how this works, i assumed dmabuf-wayland goes directly to dcp
12:41jannaudmabuf-wayland is either direct-scanout or nv12 to rgb conversion in the compositor, gpu/gpu-next is yuv2rgb in mpv
12:41chadmednot yet :P
12:42chadmedkwin does not advertise a scanout tranche for dcp yet, but it will hopefully do so Soon
12:42sofusso something about yuv->rgb must be broken right?
12:43jannauJames image looks like a interpreting nv12 as fully planar yuv400
12:45jannauthe color information is scaled by factor 2 vertically
12:45jannauno, that should look different
12:45jannaunot sure actually
12:46sofusi also noticed many videos look washed out, maybe the driver is advertising something wrong about colors?
12:46sofusthe driver being avd
12:46chadmedthe washout seems to be frame-by-frame and related to the blocking artefacts ive seen with h264 on certain videos
12:46lynndotpy72 has quit [Quit: bye bye]
12:47lynndotpy72 has joined #asahi-dev
12:47sofusmy pr should fix that, i mostly saw this with 10bit vp9
12:49voidah has quit [Ping timeout: 480 seconds]
12:49jannauI don't understand this error. looks more like the driver produces 4:2:2 but it's interpreted as 4:2:0
12:51voidah has joined #asahi-dev
12:55amarioguy has joined #asahi-dev
13:01chadmedit does the same thing for P010
13:04svendbz abridged and frieren. looks like our tastes are similar ;)
13:05sofusthe instructions dont change, so its not because avd does something different
13:05jannauit's still weird though as I doubt that image as just approximately 2 colours
13:05chadmedaah i cant grab an apothecary diaries or hellsing screenshot too, theyre both AV1 :P
13:06chaos_princesssorry if it is a stupid question, but where is the 2nd chroma channel? Isn't it supposed to be y pb pr, and i see y and pb, but no pr?
13:08jannauif the first chroma channel height lines (instead of the expected height/2) then there is just 1 chroma channel
13:08JamesCalligeros[m]let me interject for a moment what youre calling ypbpr is actually y'cbcr etc etc
13:09JamesCalligeros[m]if avd is spitting out 4:2:2 then only the horizontal resolution is subsampled but idk why then cr looks like its scaled vertically
13:09JamesCalligeros[m]anyway, they should look like
13:09sofuswith only red/black the output is green
13:12jannauyes, looks like interpreting 4:2:2 data as if it is 4:2:0
13:16chadmedyeah i wonder why dmabuf-wayland is doing the right thing then
13:16chadmedits not like it knows the buffer is nv16, the playback stats still say nv12
13:17jannauunsure where the error could be though. yuv handling for asahi/hk in mesa is afaik mostly generic shader code
13:17chadmedi suspect mpv's shaders given ff and gstreamer seem to work fine?
13:18chadmedit's not the libva translation layer, it happens with -hwdec=v4l2request too
13:18jannauthis has to happen after the decoder because no other code would be able to handle 422 data correctly when it expects 420
13:20jannaulibva produces text trace files with LIBVA_DEBUG={wayland,gpu} using the value as prefix
13:21jannaua little annoying to read since there will be multiple files (one per thread) but the main decoding should run from a single thread
13:26jannautraces are (word-)diffable, especially when decoding the same video stream with the same program
13:27jannaudifference just due to the different vo are unexpected
13:32Guest16250 has quit []
13:32enick_126 has joined #asahi-dev
13:36enick_126 has quit []
13:37ndufresne has joined #asahi-dev
13:37chadmedvalue.value.i = 0x30313050 for both vos
13:38ndufresne has quit []
13:38ndufresne has joined #asahi-dev
13:38ndufresne is now known as Guest16530
13:52Guest16530 is now known as ndufresne
14:03voidah has quit [Ping timeout: 480 seconds]
14:04ndufresne has quit [Quit: Reconnecting]
14:05ndufresne has joined #asahi-dev
14:05ndufresne is now known as Guest16534
14:08Guest16534 has quit []
14:08enick_816 has joined #asahi-dev
14:10enick_816 is now known as ndufresne
14:13voidah has joined #asahi-dev
14:13ndufresne has quit [Quit: Reconnecting]
14:14ndufresne has joined #asahi-dev
14:14ndufresne is now known as Guest16535
14:19Guest16535 is now known as ndufresne
14:21ndufresne has quit [Quit: Reconnecting]
14:21ndufresne has joined #asahi-dev
14:21voidah has quit [Ping timeout: 480 seconds]
14:21svenare device trees for the m4 pro/t6040 already floating around somewhere?
14:30voidah has joined #asahi-dev
14:35sofusit works with --gpu-api=opengl
14:36sofusalso with --hwdec={vaapi,v4l2request}-copy
14:38voidah has quit [Ping timeout: 480 seconds]
14:41svenyuka: nice, guess i can start with hv bringup directly then. does usb device mode work already?
14:42yukasven: yes
14:42svenawesome
14:43yukaidle=nop arm64.nowfxt
14:43yukaIt didn't get any reactions, maybe because it was outside of the merge window but I don't know how this works
14:44svenThank, I’m probably not gonna bother with Linux for a while then, just figured that would be a nice warm up ;)
14:44Calandracas_ has quit [Ping timeout: 480 seconds]
14:44svendepends on the subsystem
14:45svenwhat’s common is that nothing will get picked up but many still review patches from what I can tell
14:45j`eyyuka: sometimes they just pick them up later without saying anything in the mean time
14:46svenwe should probably add that stuff to the command line from m1n1 automatically
14:46yukayes, but probably only once it's upstream (or at least approved and cherry-picked in asahi-wip)
14:46yukaLooking forward to the hypervisor stuff, but publishing it will still be blocked by that publication right?
14:47crabbedhaloablut has quit []
14:47svenit‘s likely it’ll take me more time than that anyway
14:47voidah has joined #asahi-dev
14:49crabbedhaloablut has joined #asahi-dev
14:50svenwhich macOS do I want to install, is latest 27rc fine?
15:06garrettgr has quit [Ping timeout: 480 seconds]
15:14yukalatest 26.6 or 27beta/rc should both be fine
15:15yuka27 might enable more exclaves stuff, so maybe for the hypervisor work 26.6 would be easier
15:15svenack, will install the latest 26.6 then
15:15sven26.6.2 i think
15:17yukaI think we didn't know any remaining blockers for linux on 26.6.2 for M4 & Pro & Max
15:18svenalso, why did they move the power button 😫
15:18yukayou are just ~holding~ placing it wrong
15:18yukaplace it upside down and everything is fine
15:21svenI should probably solder wires to the power button and hook it up to a raspberry pico or something similar
15:22svenhahahaha :D
15:58svenTTY> CPU init (MIDR: 0x611f0551 smp_id:0x4)...
15:58svenTTY> CPU part: 0x55 rev: 0x11
15:58svenTTY> CPU: M4 Pro Brava Chop (P core)
15:58svenand all secondaries come up
16:00yukaSame on Flokli's
16:01svenhuh, mrs(IPI_RR_GLOBAL_EL1) already triggers an exception
16:03yukaThat triggers an exception on t8132 too but I didn't hit it yet when running Linux under the hv?
16:04sventhat's the AIC reg that used to be locked, isn't it?
16:04yukathat's SYS_IMP_APL_VM_TMR_FIQ_ENA_EL2
16:04svenoh... yeah
16:04sveni'm just dumb
16:04svenyeah, nvm, both mrs/msr work
16:04yukawhat did you change?
16:05yukaor you mean with the other reg
16:05yukado we want/need IPI_RR_GLOBAL_EL1 though?
16:05sveni mean with the other reg
16:05sventhat RR thing used to be fastipi but it looks like it just doesn't exist anymore?
16:06svennope, i'm just very dumb
16:06sventhat sysreg is write-only
16:06sven>>> msr(IPI_RR_LOCAL_EL1, 0)
16:06svenTTY> Exception: FIQ (from EL2h) cnt: 70b5a2404b
16:06svenTTY> Fast IPI IRQ, clearing
16:06svenwriting to the _GLOBAL works as well
16:07chaos_princesssven: mind testing if the nvram spi still works?
16:08svendo you have a script or experiment i can run?
16:09chaos_princessnot really for that one
16:10sveni haven't booted linux yet and don't really plan on doing so for now fwiw
16:11chaos_princessi just noticed that the adt on m4 lists qspi for that, but on some other machines the same node morphs between spi and qspi depending on macos ver, so want to know if the old registers still exist on m4
16:11sveni'll let you know once the hv works, should just be weeks^Hmonths :D
16:13chaos_princessQ7 2028, got it
16:14Sobek[m]I do want to give you my congrats, Sven, for that confirmed working HV in the latest blogpost.
16:14svenah crap, i forgot disabled exclaves before i installed the m1n1 boot object
16:14sven*sigh*
16:14sven*forgot to disable
16:15svensobek[m]: thanks! it only works on m1 though, now let's how hard it will be to adjust this all to m4
16:17NickChan[m]...what does that even mean? There's no GL in raw boot mode
16:18NickChan[m]about exclaves disable
16:18svenit means running some magic command that makes macos userspace or whatever not expect exclaves to be running
16:25svenhm.. can't find it anymore in the logs, but someone mentioned it in here
16:28WindowPain has quit [Ping timeout: 480 seconds]
16:30WindowPain has joined #asahi-dev
16:30chaos_princessexclaves are m5 & neo, no?
16:30svenhuh, weren't they m4 already? maybe i just misremember
16:31yukaCAEnableMedina is the keyword you want to search for
16:31yukaAnd yes M4 needs it
16:32svenah, there it is!
16:32sven16:00 <mmediouni[m]> for now, defaults write "com.apple.coreanimation" CAEnableMedina -bool false disables the exclaves code path on M5
16:37svenhm, there's no exclavecore for the m4 pro in the ipsw
16:37svenso maybe it doesn't have those yet
16:38svenexclavecore_bundle.t6050.RELEASE exclavecore_bundle.t8140.RELEASE exclavecore_bundle.t8142.RELEASE <-- that's all that ipsw tool extracted
16:38yukaMaybe in macos 27?
16:38svencould be, yeah
16:38yukaPretty sure I saw exclave things on the base M4 adt
16:39svensptm.t6041.release exists so i'd expect the exclave thingy to also be t6041
16:39WindowPain5 has joined #asahi-dev
16:43rhysmdnz has quit [Quit: Bridge terminating on SIGTERM]
16:43Guest16354 has quit [Quit: Bridge terminating on SIGTERM]
16:44WindowPain has quit [Ping timeout: 480 seconds]
16:44WindowPain5 is now known as WindowPain
16:48Jamie has joined #asahi-dev
16:48Jamie is now known as Guest16543
16:49rhysmdnz has joined #asahi-dev
17:15voidah has quit [Ping timeout: 480 seconds]
17:23voidah has joined #asahi-dev
17:35voidah has quit [Ping timeout: 480 seconds]
17:43voidah has joined #asahi-dev
17:51voidah has quit [Ping timeout: 480 seconds]
17:53sven… and the first attempt with the m1 code just directly reboots before I can even see the exception :D
17:56yukawas the m1 already running with sptm and emulated gxf/sprr?
17:56yukaor just one of them at a time
17:59voidah has joined #asahi-dev
17:59sventhey can’t really be separated
18:00yukawhere I asked if this was hardware or emulated gxf and you said hardware
18:01svenand that’s still hardware
18:01svenonce the SPRR/GXF emu is on it says „cursed hypervisor“ there
18:01yukaso this was my question: did you ever get Mac OS 27 (with sptm) to work on the "cursed" HV on M1?
18:02svenand the answer is still yes
18:02sventhat was a few days later
18:02yukaokay, thanks for confirming
18:08voidah has quit [Ping timeout: 480 seconds]
18:15okb_ has quit [Quit: Leaving]
18:16voidah has joined #asahi-dev
18:24voidah has quit [Ping timeout: 480 seconds]
18:29roxfan2 has joined #asahi-dev
18:34voidah has joined #asahi-dev
18:36roxfan has quit [Ping timeout: 480 seconds]
18:38Stary has quit [Ping timeout: 480 seconds]
18:51voidah has quit [Ping timeout: 480 seconds]
18:52zumi has joined #asahi-dev
18:52voidah has joined #asahi-dev
18:56Stary has joined #asahi-dev
19:29Stary has quit [Ping timeout: 480 seconds]
19:40Stary has joined #asahi-dev
19:52alicela1n has joined #asahi-dev
20:00nickchan has quit [Quit: ZNC 1.9.1+deb2+b3 - https://znc.in]
20:00nickchan has joined #asahi-dev
20:07___nick___ has quit [Remote host closed the connection]
20:16ddxtanx has quit [Remote host closed the connection]
20:19Calandracas has joined #asahi-dev
21:20ddxtanx has joined #asahi-dev
21:36alicela1n has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
21:43alicela1n has joined #asahi-dev
21:49alicela1n has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
21:56ivoszbg[m]stuttered a bit with ans crashing on larger writes but got it working ish
21:56ivoszbg[m]* on larger reads/writes but
21:57ivoszbg[m]also i upgraded to 27 beta 7, will check and report if the smp issue is fixed on my unit
21:58kaazoo has joined #asahi-dev
21:58voidah has quit [Remote host closed the connection]
21:59voidah has joined #asahi-dev
22:01kaazoo has quit []
22:05zumi_ has joined #asahi-dev
22:08pb17 has quit [Ping timeout: 480 seconds]
22:09zumi has quit [Ping timeout: 480 seconds]
22:10zumi_ is now known as zumi
22:10alicela1n has joined #asahi-dev
22:22pb17 has joined #asahi-dev
22:31mripard has joined #asahi-dev
22:36mripard_ has quit [Ping timeout: 480 seconds]
22:37mripard_ has joined #asahi-dev
22:43mripard has quit [Ping timeout: 480 seconds]
22:46badravixyz[m] has joined #asahi-dev
23:06voidah has quit [Ping timeout: 480 seconds]
23:14voidah has joined #asahi-dev
23:15rzhikharevich has quit [Remote host closed the connection]
23:23rzhikharevich has joined #asahi-dev
…