#
✕ nick:jannau
#asahi-dev · 2026-01-07 19:58
jannaualso there will be large overlap. the code will essentially do the same
#asahi-dev · 2026-01-07 19:57
jannauva-api or vk video doesn't really makes much of difference. Having a working driver certainly beats discussions about which api is the correct one
#asahi-dev · 2026-01-07 19:54
jannauvup: the driver for the video decoding hardware (avd) will be more on the complex side of drivers. do you have any relevant background for that or would you start completely from scratch
#asahi-dev · 2026-01-07 19:49
jannauleio: did you manage to do any work on avd?
#asahi-dev · 2026-01-07 11:12
jannauoh, there is scripts/dtc/dt_to_config in the kernel. that might be useful and looks worth testing (and fixing if there's something missing)
#asahi-dev · 2026-01-06 23:24
jannaualso suspend/resume while hdmi on macbook pros is active is fixed in asahi-wip
#asahi-dev · 2026-01-06 23:19
jannauespecially for the type-c ports it would have been nice if we had direct access to the aux channel
#asahi-dev · 2026-01-06 23:18
jannauthere are other commands to route ddc through
#asahi-dev · 2026-01-06 23:17
jannauwe have to go through one of the afk endpoints. avserv iirc. the same one which is currently used to get the EDID
#asahi-dev · 2026-01-06 23:08
jannaubacklight tries to avoid doing a commit to update the brightness so it waits to see if there's a commit/page flip in a reasonable time before it commits itself
#asahi-dev · 2026-01-06 23:07
jannauJamesCalligeros[m]: backlight updates might be faster in asahi wip. it was lovking way too many modeset locks. but I suspect you're complaining about the lazy updates
#asahi-dev · 2026-01-06 23:04
jannautesting with a network adapter on 14-inch macbook pro and at least in one case the network cable wasn't plugged fully (before I started debugging)
#asahi-dev · 2026-01-06 23:02
jannaureproduces way less than 1/3 here. interestingly both ports were affected in my single verified reproduction
#asahi-dev · 2026-01-06 23:00
jannauno luck with the broken usb-c ports on initial cold/hot plug. from tipd, usb-mux, usb-switch, usb-role-switch, dwc3 every thing looks fine and identical between broken and working state
#asahi-dev · 2026-01-06 16:08
jannauI'm not sure if efi_stub causes issue w.r.t to symbols, it doesn't for scripts/faddr2line
#asahi-dev · 2026-01-06 16:03
jannauccache could help as well
#asahi-dev · 2026-01-06 16:03
jannaualso you need to do it only once otherwise it should be most incremental compilation of 1-2 object
#asahi-dev · 2026-01-06 16:02
jannauyes. do you compile the kernel with DEBUG_INFO? and not as EFI_STUB?
#asahi-dev · 2026-01-06 15:43
jannaunot sure how that could be an issue in linux though. maybe pmgr-misc but that would be much later
#asahi-dev · 2026-01-06 15:41
jannauno that that I can rememeber but m3 max uses different number of memory controllers for the 14 and 16 core versions
#asahi-dev · 2026-01-06 15:27
jannaupushed to asahi-wip
#asahi-dev · 2026-01-06 15:23
jannauyes, seems to work. that should be less objectionable than `for_each_child_of_node() { of_platform_device_create(); }
#asahi-dev · 2026-01-06 15:21
jannauI wonder if that's enough for us as well as mtp_hid ahas a compatible
#asahi-dev · 2026-01-06 15:20
jannauthat doesn't explictly register platform devices (ignoring the pd subdevice) but uses devm_of_platform_populate
#asahi-dev · 2026-01-06 15:13
jannaudrivers/platform/chrome/chromeos_of_hw_prober.c seems to be the best reference
#asahi-dev · 2026-01-06 09:45
jannauyes, same here
#asahi-dev · 2026-01-06 09:44
jannausven: have you started looking into the unreliable coldpug / initial hotplug of the usb-c ports in asahi-wip?
#asahi-dev · 2026-01-06 09:39
jannauit's unlikely that asahi kernel driver just works. see the amount of fw version and hw variant dependent code in the driver. so if the driver loaded and tried to use the gpu I would expect that the firmware crashes immediately
#asahi-dev · 2026-01-05 19:53
jannauhv has gdbserver support but it's also a while since I used that
#asahi-dev · 2026-01-05 19:42
jannaumight be even worth sprinkling manual writes to the uart into linux early startup code if you don't where you're hanging. iirc marcan did that for the initial bringup
#asahi-dev · 2026-01-05 14:55
jannau"right now" == "this cycle", still cleanup to do. driver is very chatty
#asahi-dev · 2026-01-05 14:51
jannauasahi driver down to "18 files changed, 1384 insertions(+)". probably upper bound of what I can submit right now. there is known dead code
#asahi-dev · 2026-01-05 11:20
jannauyou have to, it's enough that the initial version had the Signed-off-by:
#asahi-dev · 2026-01-05 10:48
jannauof / device_node bindings gone, -530 lines from dependencies
#asahi-dev · 2026-01-05 09:48
jannauthe bindings are in the 6.7k from the dependencies which still can be cleaned up
#asahi-dev · 2026-01-05 09:33
jannau76 files changed, 6728 insertions(+), 147 deletions(-)
#asahi-dev · 2026-01-05 09:31
jannaudependencies are
#asahi-dev · 2026-01-05 09:31
jannau46 files changed, 18628 insertions(+)
#asahi-dev · 2026-01-05 09:31
jannaugit diff --stat drm-rust/drm-rust-next -- drivers/gpu/drm/asahi/
#asahi-dev · 2026-01-05 00:05
jannaullvmpipe renders with 2000 fps on simpledrm but display (kwin compositing on the gpu) still stutters from time to time
#asahi-dev · 2026-01-04 23:57
jannauGPU driver on simpledrm isn't a great experience. glxgears runs with 24 fps
#asahi-dev · 2026-01-04 23:52
jannauit was f6c4bdd04b8d ("arm64: dts: apple: t8103: Add missing ps_pmp dependency to ps_gfx")
#asahi-dev · 2026-01-04 23:16
jannaubits/210-gpu from asahi-wip on 6.18.3 crashes agx in the same way. we increased the stack and we have a couple of rtkit changes. the latter shouldn't make a difference
#asahi-dev · 2026-01-04 22:40
jannauadd yourself. you hit that as well
#asahi-dev · 2026-01-04 22:23
jannauit's also what the core dwc3 role switch dowes
#asahi-dev · 2026-01-04 22:22
jannauwe also should ignore noop usb role switches in the dwc3_apple. I think that's the better alternative than preventing those in tipd like I did for asahi-wip
#asahi-dev · 2026-01-04 22:19
jannauseems to be something that broke in my v6.19-rc1 branch or a dependency on other bits branches I'm not aware of
#asahi-dev · 2026-01-04 21:32
jannaufl0_id: yes, the acntrdir handling should probably be done via midr_part_features for all m3 devices in chickens.c. I started typing that but haven't pushed anything because I don't yet know what one of the bits do
#asahi-dev · 2026-01-04 21:26
jannauI guess I have too many moving parts. I'll see that I can tests less changes at once
#asahi-dev · 2026-01-04 21:24
jannauchaos_princess: did you try starlight recently? agx crashes for me during initialization with my v6.19-rc1 standalone bits/210-gpu branch. I can't see anything interesting when comparing a trace with asahi-wip. pointers differ but look consistent

plus ›