#
…

2026-03-06

00:09rany has quit [Remote host closed the connection]
00:09rany has joined #asahi-dev
00:17mixi has quit [Ping timeout: 480 seconds]
00:25gargantua has joined #asahi-dev
00:26kitlithsome idle musing about boot flow: could m1n1 theoretically provide the dynamic bits of the device tree as a device tree overlay, to be merged with <kernel specific devicetree>? granted, the eventual goal is for incompatbilities to not happen in the first place...
00:27mixi has joined #asahi-dev
00:58ddxtanx has quit [Remote host closed the connection]
01:28john-cabaj has quit [Ping timeout: 480 seconds]
02:07gargantua has quit [Ping timeout: 480 seconds]
02:43fumnanya has quit [Ping timeout: 480 seconds]
02:51fumnanya has joined #asahi-dev
03:00fumnanya has quit [Ping timeout: 480 seconds]
03:21hexdump01 has quit [Ping timeout: 480 seconds]
03:45ddxtanx has joined #asahi-dev
03:50tobhe has joined #asahi-dev
03:57tobhe_ has quit [Ping timeout: 480 seconds]
04:00tpw_rulesi've proposed that several times and been shot down
04:06nora_ has joined #asahi-dev
04:08nora has quit [Ping timeout: 480 seconds]
04:34john-cabaj has joined #asahi-dev
05:05nora has joined #asahi-dev
05:06nora_ has quit [Ping timeout: 480 seconds]
05:14JamesCalligeros[m]youre just moving the problem. the issue is that the devicetree binding constitutes an API of sorts between the platform and the drivers. whether we update m1n1 or the devicetree source itself with a break in that API is irrelevant, youd still have broken backwards compat with kernel drivers from older kernels which cant deal with it
05:14JamesCalligeros[m]and youd still need to update m1n1
05:16nickchandevice tree is kernel-specific, but the current way they are handled makes it so that there's only one device tree for all of kernels+uboot
05:16JamesCalligeros[m]yeah but having m1n1 overlay stuff on a mostly empty DTB isnt gonna fix that
05:17JamesCalligeros[m]m1n1 still needs to overlay the correct stuff
05:17JamesCalligeros[m]and "correct stuff" is a moving target
05:17JamesCalligeros[m]so either m1n1 code needs to change when bindings change, or the dts files do
05:17JamesCalligeros[m]it is easier and less annoying to just change the dts and ship the change with the relevant kernel version
05:39fumnanya has joined #asahi-dev
05:53fumnanya has quit [Ping timeout: 480 seconds]
06:02fumnanya has joined #asahi-dev
06:13betoissues has quit [Ping timeout: 480 seconds]
06:13fumnanya has quit [Ping timeout: 480 seconds]
06:15betoissues has joined #asahi-dev
06:22ddxtanx has quit [Remote host closed the connection]
06:23fumnanya has joined #asahi-dev
06:25ddxtanx has joined #asahi-dev
06:31fumnanya has quit [Ping timeout: 480 seconds]
06:31pe has quit [Remote host closed the connection]
06:31pe has joined #asahi-dev
06:38fumnanya has joined #asahi-dev
06:47fumnanya has quit [Ping timeout: 480 seconds]
06:53kitlithare the dynamic bits pulled based on information in the dtb?
06:58JamesCalligeros[m]some stuff comes from the ADT, and m1n1 already fills some of this at runtime (see kboot.c)
06:58JamesCalligeros[m]other stuff like GPU initialisation is computed
06:58JamesCalligeros[m]but this stuff isnt the problem, its the bindings themselves
07:00chadmedlets say that we wrote a driver that looks for apple,foo-prop in the devicetree. for whatever reason, that changes to apple,bar-prop in the next kernel version
07:00chadmedit doesnt matter if m1n1 is filling that or if its compiled into the DTB, the old kernel driver is expecting apple,foo-prop and doesnt find it
07:01chadmedonce devicetree bindings are upstream, they are guaranteed stable and this stops being a problem
07:03chaos_princessgpu is a pretty good example of just how much the binding can change between what we started with and what gets upstream
07:05chadmedyeah and like i said before its just better for everyone to ship devicetrees with the kernel that are guaranteed to work _with that kernel_
07:06chadmedmoving the problem into m1n1 just introduces another thing that has to constantly be updated and tracked by users. its silly
08:11pb17 has quit [Ping timeout: 480 seconds]
08:14bjoto has quit [Read error: Network is unreachable]
08:14bjoto has joined #asahi-dev
08:19ravikant_ has joined #asahi-dev
08:25pb17 has joined #asahi-dev
08:33fumnanya has joined #asahi-dev
08:58[KongrooParadox] has joined #asahi-dev
09:01KongrooParadox has quit [Ping timeout: 480 seconds]
09:08oliverI think i can configure the interchange format to only be used with external buffers, which - based on the name - might be what we want.
09:09olivermay assumption here being that external referes to something you would want to share for scanout.
09:16oliverah nevermind
09:18fumnanya has quit [Read error: Connection reset by peer]
09:18chadmedim finishing up my M1 install tonight
09:18fumnanya has joined #asahi-dev
09:19chadmedif it works fine on G13G then i dont see why we cant just allow it to be used everywhere (YCC luma channel thing notwithstanding)
09:20chadmedim more concerned about the cursor lol
09:27fumnanya has quit [Read error: Connection reset by peer]
09:27fumnanya has joined #asahi-dev
09:29oliverdo you still see issues with the cursor with the latest version? I could not reproduce the problems.
09:29oliverI would push some mesa changes though
09:32oliverI am still looking if i can get mesa to not select the interchange format for window surfaces. currently it creates those with the same modifier set as it created the screen surface. I would prefer gpu tiled in for those and i would love it to only select interchange for the screen
09:34chadmedi dont think it really matters as long as the gpu can deal with them
09:34chadmedbut yes the cursor fbs are still not being created
09:37hdbngr has joined #asahi-dev
09:38oliverAre they normally being created for you? At least gnome was always using the "software" cursor
09:38chadmedyes, we have had hardware cursors on kde for a while now
09:39chadmedgnome devs refused to support them for this platform
09:39oliverWell thanks for nothing @gnome
09:39oliverDoes it work if you run with stock mesa, but with the kernel patch?
09:44jannauoliver: the scanout buffers are either allocated on the GPU or are linear. the KMS device can only allocate dumb buffers (meaning linear, CPU accessible)
09:45oliverIs that what external means?
09:46hdbngr has quit [Ping timeout: 480 seconds]
09:46jannauinterchange should be avoided for GPU internal purposes
09:47jannauexternal buffer probably means allocated on a different device (or dma_heap)
09:52olivergot it.
09:52hdbngr has joined #asahi-dev
09:52oliverYea I'll continue to look into how to tell mesa not to use interchange for window surfaces.
09:58chadmedif the compositor wants to export the window surface to an overlay plane then it needs to
09:58chadmedmpv on kwin with KWIN_USE_OVERLAYS=1 is a good test of that
10:03chadmedi think it should be enough to just prioritise the AGX formats in mesa where we're selecting a format
10:04chadmedthat way interchange will only be selected when it is required to export a buffer to dcp/import from avd (when the time comes)
10:05oliverno does not look like that. I tried that and it reads the modifiers from some wayland dmabuf feedback thing. There it gets the interchange modifier again.
10:07chadmedi dont know if being tricky and trying to prevent userspace from doing stuff is a good idea here though beyond "if this buffer is not shared, dont select interchange"
10:08chadmedtrying to do it by content type somehow will artificially limit the compositor's choices when building a scene, and id prefer we didnt do that unless there is a hardware limitation (e.g. dcp crashing when the cursor clips the edge of the screen)
10:09hdbngr has quit [Ping timeout: 480 seconds]
10:11chadmedcursors are back with linear fbs btw (no mesa patch applied)
10:11chadmedfor some reason its all very slow though lol, somethings busted
10:12jannauoliver: it might be actually the correct thing as it is required for direct scan-out of windows
10:12chadmedwayland surfaces being interchange isnt pathological yeah
10:12chadmedthe compositor should be trying to offload them to a plane if it can
10:13jannauoliver: does gnome/mutter report a scanout tranche for zwp_linux_dmabuf_v1? check with wayland-info
10:18hdbngr has joined #asahi-dev
10:21oliverIt shows zwp_linux_dmabuf_v1, 0x0c00000000000003 is interchange
10:26hdbngr has quit [Ping timeout: 480 seconds]
10:29pb17 has quit [Ping timeout: 480 seconds]
10:45pb17 has joined #asahi-dev
10:53hdbngr has joined #asahi-dev
11:02hdbngr has quit [Ping timeout: 480 seconds]
11:03hdbngr has joined #asahi-dev
11:09jannauwhy does it report interchange for YUV formats? that looks like a bug in either mutter or mesa or a limitation in the method how the valid format / modifier combinations are queried
11:12hdbngr has quit [Ping timeout: 480 seconds]
11:12jannauquestionable if the GPU modifier work for YUV but possible as those support 1 and 2 as element size. kwin does the same with P010 and NV12
11:15hdbngr has joined #asahi-dev
11:22gargantua has joined #asahi-dev
11:22chadmedthe output on my m2 max studio is quite... interesting
11:24hdbngr has quit [Ping timeout: 480 seconds]
11:25hdbngr has joined #asahi-dev
11:38hdbngr has quit [Ping timeout: 480 seconds]
11:41jannauwith interchange cursor overlay?
11:43chadmedno the output of wayland-info
11:44chadmedthe screen looks fine, the cursor just falls back to software when the framebuffer cant be created
11:44hdbngr has joined #asahi-dev
11:53hdbngr has quit [Ping timeout: 480 seconds]
11:54chaos_princess has quit [Quit: chaos_princess]
11:54hdbngr has joined #asahi-dev
12:01chaos_princess has joined #asahi-dev
12:04hdbngr has quit [Ping timeout: 480 seconds]
12:11pe has quit [Remote host closed the connection]
12:11pe has joined #asahi-dev
12:15gargantu1 has joined #asahi-dev
12:15jannauah, I thought you updated kernel and/or mesa and got a weird rendered picture
12:15gargantua has quit [Ping timeout: 480 seconds]
12:15jannaueventually our modifiers will be reported with names: https://gitlab.freedesktop.org/mesa/libdrm/-/merge_requests/454
12:18weitcis has quit [Remote host closed the connection]
12:22pb17 has quit [Ping timeout: 480 seconds]
12:22Guest4294 has quit [Remote host closed the connection]
12:23Rock has joined #asahi-dev
12:24Rock is now known as Guest4452
12:28pb17 has joined #asahi-dev
12:31hdbngr has joined #asahi-dev
12:39hdbngr has quit [Ping timeout: 480 seconds]
12:46fumnanya has quit [Ping timeout: 480 seconds]
12:53fumnanya has joined #asahi-dev
12:55hdbngr has joined #asahi-dev
13:02fumnanya has quit [Ping timeout: 480 seconds]
13:06fumnanya has joined #asahi-dev
13:11hdbngr has quit [Ping timeout: 480 seconds]
13:16KongrooParadox has joined #asahi-dev
13:18[KongrooParadox] has quit [Ping timeout: 480 seconds]
13:26john-cabaj has quit [Ping timeout: 480 seconds]
13:26hdbngr has joined #asahi-dev
13:35hdbngr has quit [Ping timeout: 480 seconds]
13:37[KongrooParadox] has joined #asahi-dev
13:40KongrooParadox has quit [Ping timeout: 480 seconds]
13:43hdbngr has joined #asahi-dev
13:44fumnanya has quit [Ping timeout: 480 seconds]
13:51fumnanya has joined #asahi-dev
13:52hdbngr has quit [Ping timeout: 480 seconds]
13:56fumnanya has quit [Remote host closed the connection]
13:56hdbngr has joined #asahi-dev
13:57KongrooParadox has joined #asahi-dev
13:59[KongrooParadox] has quit [Ping timeout: 480 seconds]
14:00fumnanya has joined #asahi-dev
14:06noopwafelhmm, dockchannel in m1n1 breaks my macos boot
14:10hdbngr has quit [Ping timeout: 480 seconds]
14:24noopwafelah, there are two copies of the enum..
14:24chaos_princesskb/m dockchannel or serial dockchannel?
14:25lynndotpy7 has quit [Quit: bye bye]
14:26lynndotpy7 has joined #asahi-dev
14:28noopwafelthe serial one (recently added to m1n1 main). (I PRed a fix as #553)
14:37[KongrooParadox] has joined #asahi-dev
14:39KongrooParadox has quit [Ping timeout: 480 seconds]
14:54fumnanya has quit [Ping timeout: 480 seconds]
14:54hdbngr has joined #asahi-dev
15:04KongrooParadox has joined #asahi-dev
15:06hdbngr has quit [Ping timeout: 480 seconds]
15:07[KongrooParadox] has quit [Ping timeout: 480 seconds]
15:14[KongrooParadox] has joined #asahi-dev
15:15KongrooParadox has quit [Ping timeout: 480 seconds]
15:21KongrooParadox has joined #asahi-dev
15:24[KongrooParadox] has quit [Ping timeout: 480 seconds]
15:31svenwhoops, nice catch
15:37[KongrooParadox] has joined #asahi-dev
15:38pb17 has quit [Ping timeout: 480 seconds]
15:38nnym has quit [Remote host closed the connection]
15:38nnym has joined #asahi-dev
15:38pb17 has joined #asahi-dev
15:39KongrooParadox has quit [Ping timeout: 480 seconds]
15:39hightower2 has joined #asahi-dev
15:42nnym_ has joined #asahi-dev
15:49rodmac has joined #asahi-dev
15:50nnym has quit [Ping timeout: 480 seconds]
15:56hightower3 has joined #asahi-dev
15:58KongrooParadox has joined #asahi-dev
15:58hdbngr has joined #asahi-dev
16:00[KongrooParadox] has quit [Ping timeout: 480 seconds]
16:02fumnanya has joined #asahi-dev
16:03hightower2 has quit [Ping timeout: 480 seconds]
16:07hdbngr has quit [Ping timeout: 480 seconds]
16:20fumnanya has quit [Ping timeout: 480 seconds]
16:27john-cabaj has joined #asahi-dev
16:27[KongrooParadox] has joined #asahi-dev
16:29KongrooParadox has quit [Ping timeout: 480 seconds]
16:31rodmac has quit [Quit: leaving]
16:37hdbngr has joined #asahi-dev
16:47crombie has quit [Ping timeout: 480 seconds]
16:47hdbngr has quit [Ping timeout: 480 seconds]
16:50KongrooParadox has joined #asahi-dev
16:53[KongrooParadox] has quit [Ping timeout: 480 seconds]
16:59hdbngr has joined #asahi-dev
17:05pb17 has quit [Ping timeout: 480 seconds]
17:08hdbngr has quit [Ping timeout: 480 seconds]
17:12pb17 has joined #asahi-dev
17:24hdbngr has joined #asahi-dev
17:32hdbngr has quit [Ping timeout: 480 seconds]
17:33john-cabaj has quit [Read error: Connection reset by peer]
17:40hdbngr has joined #asahi-dev
17:49noshirtedkeeper[m] has joined #asahi-dev
17:49hdbngr has quit [Ping timeout: 480 seconds]
17:57michaelros4 has quit [Quit: The Lounge - https://thelounge.chat]
17:58michaelros4 has joined #asahi-dev
18:00hdbngr has joined #asahi-dev
18:01[KongrooParadox] has joined #asahi-dev
18:03KongrooParadox has quit [Ping timeout: 480 seconds]
18:05okb has joined #asahi-dev
18:09gargantua has joined #asahi-dev
18:09hdbngr has quit [Ping timeout: 480 seconds]
18:10gargantu1 has quit [Ping timeout: 480 seconds]
18:11KongrooParadox has joined #asahi-dev
18:13noopwafelso 14.8.4 is out and there's no kdk (sigh), is there any way to install 14.8.3 without wiping the whole target?
18:14[KongrooParadox] has quit [Ping timeout: 480 seconds]
18:14chaos_princesswhat do you have installed rn?
18:16noopwafel13.5, 26.x and linux (plus an empty apfs)
18:16chaos_princessrun the 14.8.4 installassistant from the 13.5 install
18:17noopwafeloooh. let me try that
18:19[KongrooParadox] has joined #asahi-dev
18:21KongrooParadox has quit [Ping timeout: 480 seconds]
18:35hdbngr has joined #asahi-dev
18:36KongrooParadox has joined #asahi-dev
18:38[KongrooParadox] has quit [Ping timeout: 480 seconds]
18:44hdbngr has quit [Ping timeout: 480 seconds]
18:55leiochadmed: oliver: there was no refusal from gnome, instead there were technical aspects that need work to make it fit in with what they are doing and no-one who has done such work
19:01hdbngr has joined #asahi-dev
19:04[KongrooParadox] has joined #asahi-dev
19:06KongrooParadox has quit [Ping timeout: 480 seconds]
19:10hdbngr has quit [Ping timeout: 480 seconds]
19:14ravikant_ has quit []
19:16hdbngr has joined #asahi-dev
19:16KongrooParadox has joined #asahi-dev
19:19[KongrooParadox] has quit [Ping timeout: 480 seconds]
19:24hdbngr has quit [Ping timeout: 480 seconds]
19:25hdbngr has joined #asahi-dev
19:29[KongrooParadox] has joined #asahi-dev
19:32KongrooParadox has quit [Ping timeout: 480 seconds]
19:34hdbngr has quit [Ping timeout: 480 seconds]
19:36KongrooParadox has joined #asahi-dev
19:38[KongrooParadox] has quit [Ping timeout: 480 seconds]
19:44laterafter[m] has joined #asahi-dev
19:46hdbngr has joined #asahi-dev
19:47[KongrooParadox] has joined #asahi-dev
19:50KongrooParadox has quit [Ping timeout: 480 seconds]
19:56KongrooParadox has joined #asahi-dev
19:57[KongrooParadox] has quit [Ping timeout: 480 seconds]
20:00hdbngr has quit [Ping timeout: 480 seconds]
20:02hdbngr has joined #asahi-dev
20:04okb has quit [Ping timeout: 480 seconds]
20:07noopwafelchaos_princess: thanks by the way, that worked.
20:13hdbngr has quit [Ping timeout: 480 seconds]
20:28[KongrooParadox] has joined #asahi-dev
20:31KongrooParadox has quit [Ping timeout: 480 seconds]
20:32fumnanya has joined #asahi-dev
20:50KongrooParadox has joined #asahi-dev
20:52[KongrooParadox] has quit [Ping timeout: 480 seconds]
21:02[KongrooParadox] has joined #asahi-dev
21:03hdbngr has joined #asahi-dev
21:04KongrooParadox has quit [Ping timeout: 480 seconds]
21:07sammy has joined #asahi-dev
21:08sammy is now known as Guest4489
21:12hdbngr has quit [Ping timeout: 480 seconds]
21:16okb has joined #asahi-dev
21:16hdbngr has joined #asahi-dev
21:21devadri94 has joined #asahi-dev
21:21devadri94 has left #asahi-dev [#asahi-dev]
21:24hdbngr has quit [Ping timeout: 480 seconds]
21:30Guest4489 has quit [Quit: Guest4489]
21:34devadri has joined #asahi-dev
21:37devadri1 has joined #asahi-dev
21:41Calandracas has joined #asahi-dev
21:44Calandracas_ has quit [Ping timeout: 480 seconds]
21:44KongrooParadox has joined #asahi-dev
21:45devadri has quit [Ping timeout: 480 seconds]
21:45devadri1 is now known as devadri
21:47[KongrooParadox] has quit [Ping timeout: 480 seconds]
21:52john-cabaj has joined #asahi-dev
22:04hdbngr has joined #asahi-dev
22:12gargantua has quit [Ping timeout: 480 seconds]
22:13hdbngr has quit [Ping timeout: 480 seconds]
22:30hdbngr has joined #asahi-dev
22:33john-cabaj has quit [Remote host closed the connection]
22:35john-cabaj has joined #asahi-dev
22:39hdbngr has quit [Ping timeout: 480 seconds]
22:39Barcher has joined #asahi-dev
22:42john-cabaj has quit [Remote host closed the connection]
22:52hdbngr has joined #asahi-dev
23:00hdbngr has quit [Ping timeout: 480 seconds]
23:01Barcher has quit [Ping timeout: 480 seconds]
23:05hdbngr has joined #asahi-dev
23:14hdbngr has quit [Ping timeout: 480 seconds]
23:19LICIA[m]1 is now known as LICIAeseiker[m]
23:48dcavalca85697976678589 has quit []
23:49dcavalca85697976678589 has joined #asahi-dev
23:53pb17 has quit [Ping timeout: 480 seconds]
23:55hdbngr has joined #asahi-dev
23:56turboretor has joined #asahi-dev
23:58pb17 has joined #asahi-dev
…