#asahi-dev · 2026-02-10 20:59
jannauI've pushed asahi-6.18.9-1 to github#asahi-dev · 2026-02-08 20:43
jannaucompatible = "apple,t6030-aic3", "apple,t8122-aic3";#asahi-dev · 2026-02-08 19:22
jannaufl0_id__: the aic3 compatible changed in asahi-wip#asahi-dev · 2026-02-08 19:22
jannaumaybe there is no need for per device dp tunables. the standalone DP phys don't have tunables either#asahi-dev · 2026-02-08 19:21
jannaunoopwafel: great. you're running out of non complex firmware devices. Is anything "simple" (as usb-c) missing?#asahi-dev · 2026-02-07 19:50
jannauhuh. why did I think there is a dedicated cd3217 for every type-c port. the even hpm%u have "usbc-fw-personality" like "HPM,35-P01-AP", "HPM,35-P23-AP" and "usbc-flash-update = 1" while the odd ones have "usbc-flash-update = 0"#asahi-dev · 2026-02-07 14:43
jannauhave you tested how fast it is? it conflicts with usb device mode cdc_acm. Is it only available on the dfu port?#asahi-dev · 2026-02-07 12:08
jannauM4 might have a native 1GHz timer#asahi-dev · 2026-02-07 12:07
jannauwriting to AGTCNTRDIR_* on M4 throws a sync exception, reading works and returns 0#asahi-dev · 2026-02-07 10:11
jannaunickchan: let's then remove the pointless #ifdef and put everything on a single line#asahi-dev · 2026-02-07 09:52
jannausigh, what an disaster is clang-format? https://github.com/AsahiLinux/m1n1/pull/543#asahi-dev · 2026-02-07 09:51
jannauFn is recognized as key: "Event: time 1770457879.914189, type 1 (EV_KEY), code 464 (KEY_FN), value 1" from evtest#asahi-dev · 2026-02-07 09:43
jannauwe do not want to expand the list of Fn mappings but use Fn as an additional modifier in user space#asahi-dev · 2026-02-07 09:33
jannaureading CNTFRQ_EL0 in EL0 traps to EL2 if BIT(0) is set#asahi-dev · 2026-02-07 09:24
jannaunickchan: I meant we need to ensure the EL1 and EL2 state is consistent for KVM#asahi-dev · 2026-02-07 09:20
jannausigh, yes. KVM needs AGTCNTRDIR_EL12. I was just thinking of m1n1 as hypervisor and there _EL1 is enough since the hypervisor loads m1n1 anyway#asahi-dev · 2026-02-07 09:04
jannauI finally found out what BIT(0) of AGTCNTRDIR does. It is an additional EL0 trap control for CNT accesses #asahi-dev · 2026-02-06 21:56
jannauI was confused by the commit date in the message and thought this was an old report. I checked that fixup was only added after that#asahi-dev · 2026-02-06 21:54
jannausven: argh, I can't read. good catch#asahi-dev · 2026-02-06 20:21
jannauand OpenCL. the GPU has a mayor quirk though. There is no preemption for compute jobs#asahi-dev · 2026-02-06 20:20
jannauthe ane driver programs DART directly instead of using linux' DMA API#asahi-dev · 2026-02-06 20:16
jannauand open user user space is not just the program which loads pre-compiled microcode but the compiler#asahi-dev · 2026-02-06 20:15
jannauhttps://docs.kernel.org/accel/introduction.html references https://docs.kernel.org/gpu/drm-uapi.html#open-source-userspace-requirements with "... All of that is the same for the accel subsystem"#asahi-dev · 2026-02-06 19:23
jannauhttps://oftc.catirclogs.org/asahi-dev/2025-09-05#asahi-dev · 2026-02-06 19:21
jannausomeone was interested in looking into the ane driver not that long ago. not sure if there anything came out of it#asahi-dev · 2026-02-05 21:30
jannauI wouldn't call usb 2.0 or charge only cables bad. what's bad is that you need a cable tester to identify cable capabilities#asahi-dev · 2026-02-05 21:01
jannau5th random cable or 5th usb3+dp-altmode / thunderbolt cable?#asahi-dev · 2026-02-05 20:59
jannauother potential issue is the cable, needs a fully connected cable (specifically connected SBU pins)#asahi-dev · 2026-02-05 20:58
jannaumacvmdtool readme says to use /dev/cu.debug_console and not /dev/tty.debug-console#asahi-dev · 2026-02-05 20:37
jannauare you using the correct usb-c port on both devices? rebooting works on any port but the serial is just on the dfu port#asahi-dev · 2026-02-05 20:35
jannausee https://asahilinux.org/docs/sw/tethered-boot/#enabling-the-backdoor-proxy-mode#asahi-dev · 2026-02-05 20:35
jannauat least I think so. my devices probably all have "-v" in boot-args and SIP disabled#asahi-dev · 2026-02-05 20:32
jannaugetz: what are you booting on the target device? if it's m1n1 there should be output at 115200 baud#asahi-dev · 2026-02-05 20:31
jannauthanks. I was remembering that someone said that#asahi-dev · 2026-02-05 20:28
jannauAppleSerialShim prevents accesing the serial port on the hsot device#asahi-dev · 2026-02-05 20:27
jannauthere might be an easier solution soon#asahi-dev · 2026-02-05 20:26
jannauit applies to the system running macvdmtool and accessing /dev/tty.debug-console on regardlessly of the os the target device is booting#asahi-dev · 2026-02-05 20:24
jannaugetz: did you see https://github.com/AsahiLinux/macvdmtool?tab=readme-ov-file#note-about-macos-12#asahi-dev · 2026-02-05 15:53
jannaukloenk: feel free to send any changes by email to me instead of a githup pr for the downstream kernel. I saw that you already did it for this change#asahi-dev · 2026-02-05 15:34
jannaukloenk: feel free to b4 send me changes. I saw the mail#asahi-dev · 2026-02-03 23:39
jannauing interesting except for m3 development. I imported the last 3 outstanding driver compatible changes to use "apple,t8103-*". spmi and watchdog important for m3, nco only when someone starts looking at audio#asahi-dev · 2026-02-03 23:36
jannauI tagged and pushed asahi-6.18.8-1 and updated asahi-wip. noth#asahi-dev · 2026-02-03 12:19
jannaufl0_id: yes, missed to adjust the m3 max entries to j514/j516. I'll fix later#asahi-dev · 2026-02-03 12:17
jannaumildsunrise has extensions for the kernel spmi controller and a description of the protocol the spmi type-c pd controller speaks (very similar or identical to the i2c one)#asahi-dev · 2026-02-03 12:14
jannauI briefly looked at usb since bringup on j504 doesn't need spmi but since the tunables differ from m1/m2 I postponed until I've macos running in the HV#asahi-dev · 2026-02-02 15:16
jannauso t603x needs to use the multi die format and 4 interrupt-cells#asahi-dev · 2026-02-02 15:16
jannauthere is a M3 Ultra (Mac Studio released last year). There probably won't be a M4 Ultra\#asahi-dev · 2026-01-31 22:16
jannauiirc I added it to boot multiple concatenated cpios#asahi-dev · 2026-01-31 22:14
jannaunoopwafel: you could add a initramfs_magic header like run_guest_kernel.sh#asahi-dev · 2026-01-31 21:51
jannaukettenis: how well tested / used is nvme write in u-boot via efi? https://codeberg.org/Limine/Limine/commit/0491f550c1b3 is reported to avoid https://github.com/AsahiLinux/linux/issues/421