#asahi-dev · 2024-03-03 12:04
jannegit allows 1.5 ms + vblank time. I would have expected that that is enough but apparently it is not#asahi-dev · 2024-03-03 12:01
jannegchadmed: kwin commits the pageflips too late for the current dcp behavior resulting in missing every second vblank and rendering at half the rate#asahi-dev · 2024-03-03 12:00
jannegleio: I haven't looked at the PR yet#asahi-dev · 2024-03-02 13:36
jannegdoesn't fix it fully but improves things#asahi-dev · 2024-03-02 13:35
jannegI'm debugging with kwin developers atm. you can try with KWIN_DRM_DISABLE_BUFFER_READABILITY_CHECKS=1 but I think you'll need https://paste.debian.net/1309235/ in addition#asahi-dev · 2024-03-02 11:09
jannegthe change is not x11 specific but it doesn't make a difference#asahi-dev · 2024-03-02 11:02
jannegpossibly related: https://bugs.kde.org/show_bug.cgi?id=481721#asahi-dev · 2024-03-02 09:51
jannegit's especially bad during window resize#asahi-dev · 2024-03-02 08:38
janneghttps://gitlab.gnome.org/GNOME/mutter/-/issues/3247#note_1973529#asahi-dev · 2024-03-02 08:38
jannegissue sounds a little bit like the issue previously reported against gnome/mutter#asahi-dev · 2024-03-02 08:29
jannegsee https://www.reddit.com/r/AsahiLinux/comments/1b3qvrw/plasma_6_runs_really_bad_under_wayland/#asahi-dev · 2024-03-02 08:27
jannegit was initially very bad (not micro stuttering) but better after an reboot#asahi-dev · 2024-03-02 08:26
jannegchadmed: I see it now (still rc2 from fedora rawhide of a couple of weeks ago), I did not notice it before#asahi-dev · 2024-02-27 20:51
jannegusing reset controllers was mentioned, I don't remember if there was a reason not to use it#asahi-dev · 2024-02-21 23:01
janneggo away, you're not welcome here#asahi-dev · 2024-02-18 08:23
jannegI doubt they share the keyboard's firmware more liekly that keychrom is emulating apple keyboards#asahi-dev · 2024-02-18 08:21
jannegthere are other fixes required for apple keyboards#asahi-dev · 2024-02-18 08:21
janneghttps://github.com/jannau/u-boot-1/tree/asahi-vidconsole-kbd-fixes#asahi-dev · 2024-02-18 08:01
jannegchadmed: the apple keyboards (2021 but probably older as well) behave the same#asahi-dev · 2024-02-17 15:33
jannegfinally have the apple keyboard working and it needs the same quirk (besides other hacks to use the interface with the boot protocol)#asahi-dev · 2024-02-17 14:40
jannegchadmed: https://paste.centos.org/view/81237712#asahi-dev · 2024-02-17 14:20
jannegthe mode make no difference here with V3. I'll have a patch. It's basically just "if (vendorId == keychron) return 1;" at the right place#asahi-dev · 2024-02-17 13:54
jannegapple usb keyboards however do not work because boot hid keyboard protocol is not on the first usb interface#asahi-dev · 2024-02-17 13:53
jannegI have a patch which quirks around this#asahi-dev · 2024-02-17 13:53
jannegas far as I understand a device issue and not u-boot's fault#asahi-dev · 2024-02-17 13:52
jannegfor the keychron keyboard user: if you press any key at the right moment during u-boot usb probing it should work. the issue is that it doesn't report anything without keypress and u-boot considers it as non-functional#asahi-dev · 2024-02-17 13:49
jannegemxll: how are you knutil invoking exactly? If you want to boot macos it looking for instructions to boot a custom kernelcache are probably more helpful#asahi-dev · 2024-02-17 11:35
jannegyes#asahi-dev · 2024-02-17 06:44
jannegthis will be annoying to trace#asahi-dev · 2024-02-17 06:43
jannegdcp might do that automatically based on swap timestamps#asahi-dev · 2024-02-17 06:40
jannegi509vcb: maybe game although games seem to have even unwanted tearing: https://www.reddit.com/r/MacOS/comments/17gg4ge/screen_tearing_since_sonoma_update/ - sonoma so might depend on the dcp firmware#asahi-dev · 2024-02-13 19:02
jannegremember to compress u-boot with gzip. that is mandatory since m1n1 can't determine size of the u-boot binary#asahi-dev · 2024-02-13 18:53
jannegdon't forget the dtbs#asahi-dev · 2024-02-13 18:52
jannegburoa: how?#asahi-dev · 2024-02-13 18:49
janneghttps://github.com/siderolabs/toolchain/blob/591759848fbd0a6b82aa1dbc3bcc9b1aa1b9fdfb/gcc/pkg.yaml#L47 might inadvertently break -nostdinc#asahi-dev · 2024-02-13 18:43
jannegand I don't see any gross hacks uin there#asahi-dev · 2024-02-13 18:40
jannegcurrent toolchain repo seems to be https://github.com/siderolabs/toolchain#asahi-dev · 2024-02-13 18:34
janneg+it#asahi-dev · 2024-02-13 18:33
jannegyes, to me as well. only open question is how did break#asahi-dev · 2024-02-13 18:32
jannegwith that adjustment the "--with-native-system-header-dir=/toolchain/include" from the gcc configure line looks suspicious as well#asahi-dev · 2024-02-13 18:29
janneg`ln -s /toolchain/include /usr/include` looks a little suspicios as well#asahi-dev · 2024-02-13 18:24
jannegor the output of `/toolchain/bin/gcc -dumpspecs`#asahi-dev · 2024-02-13 18:23
jannegthere is nothing suspicious in there. config is printed first#asahi-dev · 2024-02-13 18:20
jannegyes#asahi-dev · 2024-02-13 18:19
jannegbut I would expect the output to be the same#asahi-dev · 2024-02-13 18:19
jannegplease add -nostdinc++ instead of replacing -nostdinc#asahi-dev · 2024-02-13 18:18
jannegno, that outputs /toolchain/lib/gcc/aarch64-linux-musl/13.2.0/include#asahi-dev · 2024-02-13 18:13
jannegplease add "-v" to the command to generate display.S and paste the output from the command line#asahi-dev · 2024-02-13 18:12
jannegtry adding "-nostdinc++", that was added in gcc-13 but it would be strange if that borked gcc -nostdinc#asahi-dev · 2024-02-13 18:10
jannegfedora's 13.2.1 works as expected