#
#asahi-dev · 2026-03-11 20:20
mehdinourollahsearch/
#asahi-dev · 2026-03-11 19:58
jannauyeah, I've seen. yes, even the 26.3.2 template ADT from the IPSW still has "compatible = [wlan-pcie,bcm4387, wlan-pcie,bcm]"
#asahi-dev · 2026-03-11 18:55
nicolas17 Firmware Version: MTK_driverkit-290.14 "MTK_driverkit-290.14" Jan 27 2026 21:08:13 IO80211_driverkit-1540.16 "IO80211_driverkit-1540.16" Jan 27 2026 21:02:00
#asahi-dev · 2026-03-11 18:55
nicolas17Card Type: Wi-Fi (0x14C3, 0x7932)
#asahi-dev · 2026-03-11 18:55
nicolas17jannau: this is from a physical MacBook Neo
#asahi-dev · 2026-03-11 16:59
oliverMakes sense. And it works with xwayland
#asahi-dev · 2026-03-11 16:04
jannaunicolas17: mediatek wlan/bt for the macbook neo
#asahi-dev · 2026-03-11 15:52
leiomaybe fedora disabled a cycle earlier
#asahi-dev · 2026-03-11 15:52
leiolike the whole code to support any of that was deleted
#asahi-dev · 2026-03-11 15:52
leiooliver: GNOME upstream does not support X11 sessions anymore, only X apps via xwayland on wayland sessions
#asahi-dev · 2026-03-11 15:51
nicolas17what is "sunrise"?
#asahi-dev · 2026-03-11 13:41
noopwafelchaos_princess: good idea, thanks!
#asahi-dev · 2026-03-11 13:41
svensame
#asahi-dev · 2026-03-11 13:40
jannauI also mostly reboot before trying to remember how reload works
#asahi-dev · 2026-03-11 13:39
jannausunrise is mediatek 0x7932 closest match with linux driver is MT7922 (macbook neo)
#asahi-dev · 2026-03-11 13:37
chaos_princessbut i personally never managed to get the reloadable stuff working and always reboot
#asahi-dev · 2026-03-11 13:37
chaos_princesswatch early marcan's streams
#asahi-dev · 2026-03-11 13:35
noopwafelare there docs (or does someone have pointers) for the intended flow for writing/debugging tracers? I see all this reloadable magic so I'm wondering if there's a smarter way than rebooting every time I change something in the tracer code :)
#asahi-dev · 2026-03-11 13:05
oliverMutter has x11 disabled for fedora 43+.
#asahi-dev · 2026-03-11 08:52
chaos_princessthe msrs may just be cluster-wide
#asahi-dev · 2026-03-11 08:51
chadmedit wouldnt make sense for it to avoid the cluster at all if it wasnt at least partially shutting it off
#asahi-dev · 2026-03-11 08:51
chadmedhmm, i do know that it will try to avoid migrating tasks across clusters if there is still capacity in an already-active cluster but im not 100% sure if it shuts the unused cluster down or just puts all the cores into wfi like we do
#asahi-dev · 2026-03-11 08:49
chaos_princessnot sure, at least in oss xnu this flow is only used for full suspend
#asahi-dev · 2026-03-11 08:42
chadmeddoes macos not do this to shut off unused clusters
#asahi-dev · 2026-03-11 08:12
chaos_princessyes
#asahi-dev · 2026-03-11 08:12
chadmedso this is really only useful for platform suspend and not opportunistically sleeping clusters without some guard logic to ensure at least one core is alive
#asahi-dev · 2026-03-11 08:11
chaos_princessapple even does some complex dance where they seize all cores and run them in a spin loop to ensure nothing does a wfi before everything is ready
#asahi-dev · 2026-03-11 08:10
chadmedlmao right so if all but one core is off, and that core then happens to go into wfi without another core waking up, we're boned
#asahi-dev · 2026-03-11 08:09
yukaquestion is how much power saving we get from shutting all except one vs all
#asahi-dev · 2026-03-11 08:09
chaos_princesschadmed: according to xnu comments, the relevant msrs are global, so if the last core executes a wfi, it will also go to sleep
#asahi-dev · 2026-03-11 08:09
chadmede.g. if all the pcores shut off and the soc shuts off the uncore cluster parts, that is a nice power saving, and whatever we leave behind in m1n1 can spin them back up enough for linux to restore state
#asahi-dev · 2026-03-11 08:08
chadmedshutting down only the secondaries might be fine for a cpuidle hack, hard to say
#asahi-dev · 2026-03-11 08:08
chaos_princessthe first step `p.smp_start_secondaries(); p.smp_stop_secondaries(True)` does exactly that. It is shutting down the last core to make all supporting parts go into sleep that is interesting
#asahi-dev · 2026-03-11 08:08
chadmedmaybe, but if we want to support s2ram then linux will try to shut them all down anyway
#asahi-dev · 2026-03-11 08:06
yukachaos_princess: re sleep does this fully shut down all cores? If yes, would shutting down all cores except one E-core work?
#asahi-dev · 2026-03-11 01:42
JamesCalligeros[m]we havent figured out exactly how that works. e.g. i cannot get surface 2 to display anything yet macos can put stuff on it no worries
#asahi-dev · 2026-03-11 01:41
JamesCalligeros[m]but it also knows what to pass off to IOMFB to get all the surfaces to work properly
#asahi-dev · 2026-03-11 01:40
JamesCalligeros[m]falls back to software
#asahi-dev · 2026-03-11 01:40
nicolas17what does macOS do when the cursor goes off screen?
#asahi-dev · 2026-03-11 01:39
JamesCalligeros[m]if our planes are so fucked up that we cant do that, we cant use them
#asahi-dev · 2026-03-11 01:38
JamesCalligeros[m]we have to obey all the kms api's invariants and accept whatever the consumers of those apis give us
#asahi-dev · 2026-03-11 01:38
JamesCalligeros[m]one of the issues we have is that the macos compositor is allowed to cheat and paper over dcp's insanity
#asahi-dev · 2026-03-11 01:37
JamesCalligeros[m]yes
#asahi-dev · 2026-03-11 01:23
nicolas17does macOS use a hardware overlay for the mouse cursor?
#asahi-dev · 2026-03-10 23:59
oliverI can imagine that a gtk app can tell the compositor: hey I have actually two surfaces. Here is one fully opaque in yuv (that I got from the video decoder) and here is a transparent one. Both are passed via dmabuf fb to the wayland compositor.. The compositor can then put them onto planes and pass them to the dcp (or hardware compositor) if possible, or compose them on the gpu if not.
#asahi-dev · 2026-03-10 23:08
leiobut now I'm just still confused how all of these things relate to eachother wrt dmabuf surfaces as supposedly used by modern GTK video players vs universal planes vs compositors vs DCP
#asahi-dev · 2026-03-10 23:06
leioIn an ideal world I still like the aspect of compositors being able to use more overlay planes for other things than the cursor, being able to opt to render cursor via SW (or maybe it's disabled overall due to being on top of a video)
#asahi-dev · 2026-03-10 23:01
oliverI don't think you can do that easily in the kernel. :/
#asahi-dev · 2026-03-10 23:00
JamesCalligeros[m]im not testing xorg though someone else can do that
#asahi-dev · 2026-03-10 22:59
JamesCalligeros[m]if we can get something similar to work in the kernel we can just do it there and allocate surface 3 as a cursor plane and not break anyone or anything

plus ›