#asahi-dev · 2026-03-06 10:11
chadmedfor some reason its all very slow though lol, somethings busted#asahi-dev · 2026-03-06 10:11
chadmedcursors are back with linear fbs btw (no mesa patch applied) #asahi-dev · 2026-03-06 10:08
chadmedtrying 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) #asahi-dev · 2026-03-06 10:07
chadmedi 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"#asahi-dev · 2026-03-06 10:04
chadmedthat way interchange will only be selected when it is required to export a buffer to dcp/import from avd (when the time comes)#asahi-dev · 2026-03-06 10:03
chadmedi think it should be enough to just prioritise the AGX formats in mesa where we're selecting a format#asahi-dev · 2026-03-06 09:58
chadmedmpv on kwin with KWIN_USE_OVERLAYS=1 is a good test of that#asahi-dev · 2026-03-06 09:58
chadmedif the compositor wants to export the window surface to an overlay plane then it needs to#asahi-dev · 2026-03-06 09:40
chadmedill try#asahi-dev · 2026-03-06 09:39
chadmedgnome devs refused to support them for this platform #asahi-dev · 2026-03-06 09:38
chadmedyes, we have had hardware cursors on kde for a while now #asahi-dev · 2026-03-06 09:34
chadmedbut yes the cursor fbs are still not being created#asahi-dev · 2026-03-06 09:34
chadmedi dont think it really matters as long as the gpu can deal with them#asahi-dev · 2026-03-06 09:20
chadmedim more concerned about the cursor lol#asahi-dev · 2026-03-06 09:19
chadmedif it works fine on G13G then i dont see why we cant just allow it to be used everywhere (YCC luma channel thing notwithstanding) #asahi-dev · 2026-03-06 09:18
chadmedim finishing up my M1 install tonight#asahi-dev · 2026-03-06 07:06
chadmedmoving the problem into m1n1 just introduces another thing that has to constantly be updated and tracked by users. its silly #asahi-dev · 2026-03-06 07:05
chadmedyeah and like i said before its just better for everyone to ship devicetrees with the kernel that are guaranteed to work _with that kernel_#asahi-dev · 2026-03-06 07:01
chadmedonce devicetree bindings are upstream, they are guaranteed stable and this stops being a problem#asahi-dev · 2026-03-06 07:00
chadmedit 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#asahi-dev · 2026-03-06 07:00
chadmedlets 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#asahi-dev · 2026-03-04 07:16
chadmedjannau: sorry i did look at the CI pull request a few days ago and it looked fine to me, just forgot to comment/approve #asahi-dev · 2026-03-04 07:15
chadmedif you really think about it, the increased battery life is just like adding a bunch of fuel tanks to the XLRs :P #asahi-dev · 2026-03-03 20:48
chadmedmacbook New Engine Option? #asahi-dev · 2026-03-03 12:18
chadmedhttps://github.com/AsahiLinux/linux/pull/464#asahi-dev · 2026-03-03 08:58
chadmedjannau: do you have any objections to pulling a trivial patch to "fix" sample rate issues with cs42l84? i will submit upstream this weekend or so but im being bugged about it :P '#asahi-dev · 2026-03-02 20:50
chadmedit's not corruption, plasma just falls back to a software cursor. you can see this if you use drm_info; there is no framebuffer attached to the overlay plane#asahi-dev · 2026-03-02 20:49
chadmedcould be a kwin 6.6.1 regression then. i will do some testing after work#asahi-dev · 2026-03-02 12:43
chadmedhahaha yep firefox macos UA on linux gives a 403 too #asahi-dev · 2026-03-02 11:28
chadmedhttps://github.com/AsahiLinux/linux/pull/463#asahi-dev · 2026-03-02 10:52
chadmedalright yeah that works#asahi-dev · 2026-03-02 10:33
chadmedoh i have a gross idea#asahi-dev · 2026-03-02 10:33
chadmedactually no that doesnt work, plasma frees the CTM blob when turning night light off#asahi-dev · 2026-03-02 10:22
chadmedbecause now this has the opposite problem - it clears the existing ctm on every atomic_enable :P #asahi-dev · 2026-03-02 10:21
chadmedi wonder if theres an easy way to check if the installed CTM is not the identity already #asahi-dev · 2026-03-02 10:20
chadmedok that kinda works#asahi-dev · 2026-03-02 07:46
chadmedyeah. ill look at it after dinner#asahi-dev · 2026-03-02 06:43
chadmedmaybe if theres a colour management change committed while the crtc is off we should queue it up or something #asahi-dev · 2026-03-02 06:43
chadmedif plasma is offloading colour ops to kms it cant switch night light off if the screen is off #asahi-dev · 2026-03-02 06:42
chadmedwe need to do something about the CTM not changing if the crtc is off#asahi-dev · 2026-03-01 20:48
chadmedjannau: yeah the cursor plane is using interchange#asahi-dev · 2026-03-01 13:03
chadmedin the meantime ill just use my massive animated kasane teto cursor set :P #asahi-dev · 2026-03-01 12:55
chadmedpossibly. i couldnt find anywhere where either an atomic_check or mesa returned a new einval but its also 11pm so #asahi-dev · 2026-03-01 12:53
chadmedif you want to repro, i use 120% scaling and "size" 24 breeze cursors#asahi-dev · 2026-03-01 12:52
chadmedhopefully your travel insurance doesnt pull some "we dont cover acts of war" bullshit #asahi-dev · 2026-03-01 12:52
chadmedstay safe! #asahi-dev · 2026-03-01 12:52
chadmedoh dear#asahi-dev · 2026-03-01 12:30
chadmedbut yes looks related #asahi-dev · 2026-03-01 12:29
chadmedi dont think so sadly#asahi-dev · 2026-03-01 12:13
chadmedi shall comment on the PR too