#
✕ nick:jannau
#asahi-dev · 2026-07-01 19:24
jannauhas the ADT a misleading speaker config, is my understanding of how to map ADT to DT wrong or is our audio DT broken (14/16-inch MBP): https://github.com/AsahiLinux/m1n1/pull/612 ?
#asahi-dev · 2026-07-01 06:44
jannausofus: I tested with this patch: https://paste.debian.net/hidden/d8f48d10
#asahi-dev · 2026-06-30 17:51
jannaufilevault was enabled but the behavior did not change after disabling it. still asked to select a volume and to authenticate after disabling fv. possibly m5 specific
#asahi-dev · 2026-06-30 15:37
jannauI might have enabled "find my" without thinking/reading
#asahi-dev · 2026-06-30 15:34
jannauis macOS recovery assistant a golden gate thing? "Options" from the startup picker requested that I select a volume to recover and then asked for the admin password. Or is this filevault? If it's enabled the installer tricked me to enabling it
#asahi-dev · 2026-06-30 11:33
jannausofus: how much did you modify the rockchip (rkvdec) code? is it feasible to reuse the code? Upstream most likely will ask this as well
#asahi-dev · 2026-06-30 09:27
jannauhttps://gitlab.com/kernel-firmware/linux-firmware/-/merge_requests/833 has no source for MIT licensed firmware. I think I would prefer to submit just the binaries and refer to the fw repository
#asahi-dev · 2026-06-30 09:17
jannaunot sure how the corresponding source code should be included. I guess doesn't need be included with MITS/BSD
#asahi-dev · 2026-06-30 09:15
jannauhttps://gitlab.com/kernel-firmware/linux-firmware/-/blob/main/LICENSE-CRITERIA.md#22-standard-open-source-licenses so the firmware should go into linux-firmware
#asahi-dev · 2026-06-30 09:10
jannauonly if everything is your own. the license is for the whole file
#asahi-dev · 2026-06-30 09:09
jannauyes, if we would submit it to linux-firmware it would be the binaries only
#asahi-dev · 2026-06-30 09:07
jannauis open-source firmware typically included in linux-firmware? I guess it helps distros which don't have a cross-toolchain to build the firmware but from the m1n1 packaging I'm under the impression that Fedora prefers to build everything on its own infrastructure
#asahi-dev · 2026-06-30 09:04
jannauit can't be -or-MIT when it is based on the GPL only driver. specific parts which are not based on GPL source could be -or-MIT
#asahi-dev · 2026-06-29 12:02
jannauan option to print the adt with translated addresses would make sense for writing device-trees
#asahi-dev · 2026-06-29 12:00
jannau`python3 -m m1n1.adt -a` does the address translation
#asahi-dev · 2026-06-29 10:41
jannauintegralpilot: we can do further review also on list
#asahi-dev · 2026-06-29 10:41
jannauintegralpilot: my advice re hid country-code was bad. I remembered that we need/use it to differentiate between the ansi, iso, and jis keyboard layout variants in apple-hid
#asahi-dev · 2026-06-29 10:38
jannauintegralpilot: don't reuse USB IDs for dockchannel in the apple keyboard driver
#asahi-dev · 2026-06-29 10:37
jannauintegralpilot: I still have comments. apple-dockchannel-hid/apple-dockchannel-hid.c is excessive as folder / file name. I think dockchannel/ is distinctive enough as folder name. maybe dockchannel/apple.c or dockchannel/apple-hid.c
#asahi-dev · 2026-06-29 10:08
jannauchadmed: how can we verify the dai-tdm-slot-tx-idle-mask? I might have switched some unnecessarily. Is testing speakers on each device enough or could there be subtle issues?
#asahi-dev · 2026-06-29 10:06
jannauyuka: some would appreciate the output of https://github.com/AsahiLinux/m1n1/pull/611
#asahi-dev · 2026-06-29 10:05
jannauintegralpilot: thanks
#asahi-dev · 2026-06-29 07:40
jannauplease don't. the APIs we want to support (va-api/vulkan video) are stateless as well
#asahi-dev · 2026-06-29 07:39
jannauI think I remember a toot mentioning speakersafetyd and iirc ayn devices
#asahi-dev · 2026-06-29 07:38
jannaumac mini/studio don't have "sound-name-prefix", easily changed
#asahi-dev · 2026-06-29 07:28
jannaudo we have speakersafetyd use on non-apple devices?
#asahi-dev · 2026-06-29 07:27
jannauand both are in the bindings so we can rely on it
#asahi-dev · 2026-06-29 07:26
jannauwe can match on "sound-name-prefix"
#asahi-dev · 2026-06-29 07:24
jannauyes, I still expect the speaker config on all 14/16 inch macbook pros to be identical
#asahi-dev · 2026-06-29 07:22
jannauwould it make sense to either add a check to speakersafetyd that it conf matches the live devicetree or simply read is_chan/vs_chan directly from there?
#asahi-dev · 2026-06-29 07:19
jannauj504.conf has a 10 degree lower t_limit for the woofers. would it make sense to check if the other confs in speakersafetyd still still match macos?
#asahi-dev · 2026-06-29 07:16
jannaudoes speakersafetyd's is_chan/vs_chan is supposed to match ti,imon-slot-no/ti,vmon-slot-no (deviced by 2)? if so we can't link speakersafetyd's j514/j516 conf to j314/j316
#asahi-dev · 2026-06-28 20:54
jannau2048 is the fifo size
#asahi-dev · 2026-06-28 20:24
jannauaudio tested on j504/j514s
#asahi-dev · 2026-06-28 20:23
jannauchadmed: I didn't drop the mca shutdown commit since I did not want to deal with the conflicts
#asahi-dev · 2026-06-28 20:22
jannauI've rebased asahi-wip onto v7.1.2, not tagged and I've ignored PRs for now
#asahi-dev · 2026-06-28 12:04
jannauit doesn't because USE_PMP is not defined. we can sort this out when we decide to use PMP
#asahi-dev · 2026-06-28 12:03
jannauI have access to a m3 ultra but it's low priotrity since it's macos 15 and later
#asahi-dev · 2026-06-28 12:01
jannauoverriding ans' power-domain from t6032.dtsi isn't much of an issue, it will need anyway to add all the devices from the second die. although the sram is probably used on both dies
#asahi-dev · 2026-06-28 10:13
jannauwith v7.3 the device lifetimes and io projection are hopefully stable so that makes it a good future target as well
#asahi-dev · 2026-06-28 10:07
jannaure common code: I thought that would be taken care of via rust kbuild changes (afaik planned for v7.3). nova-{core,drm} has it's own way. not sure if that still going forward
#asahi-dev · 2026-06-28 10:03
jannauyuka: our downstream pcie-poweron changes were soft-NAKed, it should be reimplemented with PCI_PWRCTRL_GENERIC (v7.1) / https://lore.kernel.org/linux-pci/20260220-topic-sm8650-ayaneo-pocket-s2-base-v5-3-1ad79caa1efa@linaro.org/
#asahi-dev · 2026-06-28 09:56
jannauupstreaming of drivers in bits/250-aop (more like rust-drivers-except-agx) would be appreciated. It's blocked on the rtkit rust bindings which might not play well with rust device lifetime changes currently in progress. so it might not be the best target
#asahi-dev · 2026-06-28 09:51
jannauintegralpilot: I haven't looked much beyond the mailbox changes. I'm heat throtled until tomorrow
#asahi-dev · 2026-06-27 09:37
jannauchaos_princess suggested that earlier
#asahi-dev · 2026-06-27 08:50
jannaudcp can scan semi-planar surfaces as well so no issue there
#asahi-dev · 2026-06-27 08:46
jannauisn't that V4L2_PIX_FMT_NV20? see https://www.kernel.org/doc/html/latest/userspace-api/media/v4l/pixfmt-yuv-planar.html
#asahi-dev · 2026-06-27 08:37
jannauignoring interchange compressed surfaces
#asahi-dev · 2026-06-27 08:37
jannauwhat do we need which isn't present yet in v4l2
#asahi-dev · 2026-06-27 08:36
jannauI'm at a stage where this could a millionm different things. I would the combination to support nv12 just fine

plus ›