#asahi-dev · 2026-08-12 08:36
chadmedcould someone with an up to date gnome system please test https://github.com/chadmed/linux/tree/dcp/interchange along with https://gitlab.freedesktop.org/chadmed/mesa/-/tree/asahi/interchange?ref_type=heads#asahi-dev · 2026-08-12 03:53
chadmedall i want is an interchange buffer smdh #asahi-dev · 2026-08-12 03:52
chadmed[ 184.342412] apple-dcp 289c00000.dcp: RTKit: syslog message: PPipeDCP_H13P.cpp:10259: IOMFB: polarity stuck! Resetting display#asahi-dev · 2026-08-11 13:12
chadmednot sure, all ive noticed is that colours change slightly on some content between compositing and scanout#asahi-dev · 2026-08-11 13:05
chadmedi need to look at colour pipelines again soon, we do hit the colour bug mentioned in the comments on that MR under certain conditions#asahi-dev · 2026-08-11 13:04
chadmed8076 got merged so i think we should be okay from a kwin perspective to enable overlays unconditionally no#asahi-dev · 2026-08-11 13:01
chadmedhm yes since that is technically a regression#asahi-dev · 2026-08-11 12:56
chadmedactually try set KWIN_USE_OVERLAYS=1 #asahi-dev · 2026-08-11 12:56
chadmedplane 2 gets the cursor, plane 1 gets nothing without KWIN_USE_OVERLAYS, plane 0 gets the desktop#asahi-dev · 2026-08-11 12:55
chadmedim using it right now on this machine#asahi-dev · 2026-08-11 12:49
chadmeddo you have KWIN_FORCE_SW_CURSOR=1 set somewhere#asahi-dev · 2026-08-11 12:49
chadmednope, working fine on 6.7.4 here#asahi-dev · 2026-08-11 08:06
chadmednice!#asahi-dev · 2026-08-11 07:27
chadmed(other than the nuclear option of smushing the system version of course) #asahi-dev · 2026-08-11 07:27
chadmedis there a better way to test local mesa builds than meson devenv? it's kinda flakey inside muvm#asahi-dev · 2026-08-10 13:09
chadmedthis isnt causing crashes but graphical corruption. debugging the standard gaming stack is painful and tedious at times, but you could try using wined3d in the first instance #asahi-dev · 2026-08-10 13:01
chadmedproblem for another day though #asahi-dev · 2026-08-10 13:00
chadmedit falls back if descriptor heaps dont work, and that besides, 2.7 is also broken#asahi-dev · 2026-08-10 12:46
chadmedproton vendors its own version#asahi-dev · 2026-08-10 12:39
chadmedi dont use fedora/the "standard" gaming stack so not 100% sure#asahi-dev · 2026-08-10 12:39
chadmedid say so, 3.0 is recent and introduced a lot of major changes#asahi-dev · 2026-08-10 12:34
chadmeddxvk 2.7.1 is also broken#asahi-dev · 2026-08-10 11:43
chadmedgames that used to be fine with dxvk are now unplayable#asahi-dev · 2026-08-10 11:43
chadmedhas honeykrisp seriously regressed or has dxvk 3.0 introduced requirements on extensions we do not have#asahi-dev · 2026-08-10 10:49
chadmedno, you should be able to use usb #asahi-dev · 2026-08-09 06:25
chadmedwas just curious if we knew how to make avd do it#asahi-dev · 2026-08-09 06:25
chadmedyeah but not important until the gpu and dcp are sorted out#asahi-dev · 2026-08-09 06:07
chadmedoh god what a mess, looks like each vendor just creates their own pixfmts lol#asahi-dev · 2026-08-09 06:03
chadmedim not entirely sure. does v4l2 support format modifiers in the same way drm does?#asahi-dev · 2026-08-09 05:56
chadmedthere may be cases where the gpu cant read interchange (i dont think anyone tested G13 but potentially pre-M-series AGX too) but that doesnt matter for direct scanout since all the dcp variants we support can read them#asahi-dev · 2026-08-09 05:53
chadmedto jog your memory: https://oftc.catirclogs.org/asahi-dev/2026-06-22#35437355;#asahi-dev · 2026-08-09 05:53
chadmedim just thinking _potentially_ some of the artefacts/slowness could be related to copying big linear buffers to the gpu. both the gpu and dcp can read interchange #asahi-dev · 2026-08-09 05:52
chadmedno rush, still need to get a bunch of other dcp-related stuff merged first#asahi-dev · 2026-08-09 05:50
chadmedit should be using them already as the surface they end up scanned out on is an interchange surface#asahi-dev · 2026-08-09 00:44
chadmedsofus: do we know how to make AVD export interchange compressed buffers?#asahi-dev · 2026-08-08 11:33
chadmedim not sure why they used to block DP MST on intel macs that supported it just fine, alas#asahi-dev · 2026-08-08 11:32
chadmedThink Different#asahi-dev · 2026-08-08 11:31
chadmedis there even much media that actually uses film grain reconstruction at all? ive not seen any encodes that make use of it yet#asahi-dev · 2026-08-08 11:29
chadmednice#asahi-dev · 2026-08-08 08:17
chadmedtheres also that rf filter thingy hanging off aop that has something to do with the speakers and compass#asahi-dev · 2026-08-08 08:16
chadmedtheyre bringing back the headphone jack and putting in a broadcom fm superhet for broadcast radio :P #asahi-dev · 2026-08-08 08:16
chadmed"advanced RF components" could mean literally anything#asahi-dev · 2026-08-08 08:15
chadmedif theyre just building the frontend for apple's modem thats different#asahi-dev · 2026-08-08 08:08
chadmedhopefully Cx or whatever its called is at least more stable#asahi-dev · 2026-08-08 08:08
chadmedits specifically having to care about brcmfmac and the rogues gallery of unstable crappy wifi/bt chips in these machines#asahi-dev · 2026-08-08 08:07
chadmedyeah but we dont have to care about those too much and they have mtp in front of it now#asahi-dev · 2026-08-08 08:05
chadmedi think the mediatek part ended up mostly being compatible with a driver already upstream? #asahi-dev · 2026-08-08 08:05
chadmedi am aware yes#asahi-dev · 2026-08-08 08:04
chadmedmmediouni[m]: not for us :P#asahi-dev · 2026-08-08 08:04
chadmedtranslators note: rtk means rtkit (not realtek)